Task Understanding from Confusing Multi-task Data: Difference between revisions

From statwiki
Jump to navigation Jump to search
 
(62 intermediate revisions by 31 users not shown)
Line 5: Line 5:
= Introduction =
= Introduction =


Narrow AI is an artificial intelligence that outperforms humans in a narrowly defined task. The application of Narrow AI is becoming more and more common. For example, Narrow AI can be used for spam filtering, music recommendation services, and even self-driving cars. However, the widespread use of Narrow AI in important infrastructure functions raises some concerns. Some people think that the characteristics of Narrow AI make it fragile, and when neural networks can be used to control important systems (such as power grids, financial transactions), alternatives may be more inclined to avoid risks. While these machines help companies improve efficiency and cut costs, the limitations of Narrow AI encouraged researchers to look into General AI.  
Narrow AI is an artificial intelligence that outperforms humans in a narrowly defined task. The application of Narrow AI is becoming more and more common. For example, Narrow AI can be used for spam filtering, music recommendation services, assisting doctors to make data-driven decisions, and even self-driving cars. One of the most famous integrated forms of Narrow AI is Apple's Siri. Siri has no self-awareness or genuine intelligence, and hence often has challenges performing tasks outside its range of abilities. However, the widespread use of Narrow AI in important infrastructure functions raises some concerns. Some people think that the characteristics of Narrow AI make it fragile, and when neural networks can be used to control important systems (such as power grids, financial transactions), alternatives may be more inclined to avoid risks. While these machines help companies improve efficiency and cut costs, the limitations of Narrow AI encouraged researchers to look into General AI.  


General AI is a machine that can apply its learning to different contexts, which closely resembles human intelligence. This paper attempts to generalize the multi-task learning system that learns from data from multiple classification tasks. One application is image recognition. In figure 1, an image of an apple corresponds to 3 labels: “red”, “apple” and “sweet”. These labels correspond to 3 different classification tasks: color, fruit, and taste.  
General AI is a machine that can apply its learning to different contexts, which closely resembles human intelligence. This paper attempts to generalize the multi-task learning system that learns from data from multiple classification tasks. For an isolated and very difficult task, the artificial intelligence may not learn it very well. For instance, a net with pixel dimension 1000*1000 is less likely to identify complicated objects in real-world situations on the time basis. However, if it could be learned simultaneously, it would be better as the tasks can share what they learned. It is easier for the learner to learn together instead of in isolation, for example, shapes, landmarks, textures, orientation and so on. This is called Multitask Learning. One application is image recognition. In figure 1, an image of an apple corresponds to 3 labels: “red”, “apple” and “sweet”. These labels correspond to 3 different classification tasks: color, fruit, and taste.  


[[File:CSLFigure1.PNG | 500px]]
[[File:CSLFigure1.PNG | 500px]]


Currently, multi-task machines require researchers to construct a task definition. Otherwise, it will end up with different outputs with the same input value.  Researchers manually assign tasks to each input in the sample to train the machine. See figure 1(a). This method incurs high annotation costs and restricts the machine’s ability to mirror the human recognition process. This paper is interested in developing an algorithm that understands task concepts and performs multi-task learning without manual task annotations.  
Currently, multi-task machines require researchers to construct a task definition. Otherwise, it will end up with different outputs with the same input values.  Researchers manually assign tasks to each input in the sample to train the machine. See figure 1(a). This method incurs high annotation costs and restricts the machine’s ability to mirror the human recognition process. This paper is interested in developing an algorithm that understands task concepts and performs multi-task learning without manual task annotations.  


This paper proposed a new learning method called confusing supervised learning (CSL) which includes 2 functions: de-confusing function and mapping function. The first function allocates identifies an input to its respective task and the latter finds the relationship between the input and its label. See figure 1(b). To train a network of CSL, CSL-Net is constructed for representing CSL’s variables. However, this structure cannot be optimized by gradient back-propagation. This difficulty is solved by alternatively performing training for the de-confusing net and mapping net optimization.  
This paper proposed a new learning method called confusing supervised learning (CSL) which includes two functions: de-confusing function and mapping function. The de-confusing function allocates samples to respective tasks and the mapping function presents the relation from the input to its label within the allocated tasks. See figure 1(b). To implement the CSL, we use a risk functional to balance the effects of the de-confusing function and mapping function.  


Experiments for function regression and image recognition problems were constructed and compared with multi-task learning with complete information to test CSL-Net’s performance. Experiment results show that CSL-Net can learn multiple mappings for every task simultaneously and achieve the same cognition result as the current multi-task machine sigh complete information.
However, simply combining the two functions or networks to a single architecture is impossible, since the one-hot constraint of the outputs for the de-confusing network makes the gradient back-propagation unfeasible. This difficulty is solved by alternatively performing training for the de-confusing net and mapping net optimization in the proposed architecture CLS-Net.
 
Experiments for function regression and image recognition problems were constructed and compared with multi-task learning with complete information to test CSL-Net’s performance. Experiment results show that CSL-Net can learn multiple mappings for every task simultaneously and achieve the same cognition result as the current multi-task machine assigned with complete information.


= Related Work =
= Related Work =


[[File:CSLFigure2.PNG | 700px]]
[[File:CSLFigure2.PNG | 700px]]
==Latent variable learning==
Latent variable learning aims to estimate the true function with mixed probability models. See '''figure 2a'''. In the multi-task learning problem without task annotations, we know that samples are generated from multiple distinct distributions instead of one distribution, combining multiple probability models. Thus, the latent variable learning can not fully distinguish labels into different tasks and distributions, and it is insufficient to classify the multi-task confusing samples.


==Multi-task learning==
==Multi-task learning==
Multi-task learning aims to learn multiple tasks simultaneously using a shared feature representation. In multi-task learning, the task to which every sample belongs is known. By exploiting similarities and differences between tasks, the learning from one task can improve the learning of another task. (Caruana, 1997) This results in improved learning efficiency. Multi-task learning is used in disciplines like computer vision, natural language processing, and reinforcement learning. In multi-task learning, the task to which every sample belongs is known. With this task definition, the input-output mapping of every task can be represented by a unified function. However, these task definitions are manually constructed, and machines need manual task annotations to learn. Without this annotation, our goal is to understand the task concept from confusing input-label pairs. Overall, It requires manual task annotation to learn and this paper is interested in machine learning without a clear task definition and manual task annotation.
Multi-task learning aims to learn multiple tasks simultaneously using a shared feature representation. By exploiting similarities and differences between tasks, the learning from one task can improve the learning of another task. (Caruana, 1997) This results in improving the overall learning efficiency since the labels in different tasks are often correlated: improving the classification result for one class also helps with other classification tasks. In multi-task learning, the input-output mapping of every task can be represented by a unified function. However, these task definitions are manually constructed, and machines need manual task annotations to learn. If such manual task annotation is absent, then the algorithm can not be performed.
 
==Latent variable learning==
Latent variable learning aims to estimate the true function with mixed probability models. See '''figure 2a'''. In the multi-task learning problem without task annotations, samples are generated from multiple distributions instead of one distribution. Thus, latent variable learning is insufficient to solve the research problem, which is multi-task confusing samples.


==Multi-label learning==
==Multi-label learning==
Multi-label learning aims to assign an input to a set of classes/labels. See '''figure 2b'''. It is a generalization of multi-class classification, which classifies an input into one class. In multi-label learning, an input can be classified into more than one class. Unlike multi-task learning, multi-label does not consider the relationship between different label judgments and it is assumed that each judgment is independent. An example where multi-label learning is applicable is the scenario where a website wants to automatically assign applicable tags/categories to an article. Since an article can be related to multiple categories (eg. an article can be tagged under the politics and business categories) multi-label learning is of primary concern here.
Multi-label learning aims to assign an input to a set of classes/labels. See '''figure 2b'''. It is a generalization of multi-class classification, which classifies an input into one class. In multi-label learning, an input can be classified into more than one class. Unlike multi-task learning, multi-label does not consider the relationship between different label judgments, and it is assumed that each judgment is independent. An example where multi-label learning is applicable is the scenario where a website wants to automatically assign applicable tags/categories to an article. Since an article can be related to multiple categories (e.g., an article can be tagged under the politics and business categories), multi-label learning is of primary concern.


= Confusing Supervised Learning =
= Confusing Supervised Learning =
Line 34: Line 36:
== Description of the Problem ==
== Description of the Problem ==


Confusing supervised learning (CSL) offers a solution to the issue at hand. A major area of improvement can be seen in the choice of risk measure. In traditional supervised learning, let <math> (x,y)<math> be the training samples from <math>y=f(x)</math>, which is an identical but unknown mapping relationship. Assuming the risk measure is mean squared error (MSE), the expected risk functional is
Confusing supervised learning (CSL) offers a solution to the issue at hand. A major area of improvement can be seen in the choice of risk measure. In traditional supervised learning, let <math> (x,y)</math> be the training samples from <math>y=f(x)</math>, which is an identical but unknown mapping relationship. Assuming the risk measure is mean squared error (MSE), the expected risk function is


$$ R(g) = \int_x (f(x) - g(x))^2 p(x) \; \mathrm{d}x $$
$$ R(g) = \int_x (f(x) - g(x))^2 p(x) \; \mathrm{d}x $$


where <math>p(x)</math> is the prior distribution of the input variable <math>x</math>. In practice, model optimizations are performed using the empirical risk
where <math>p(x)</math> is the data distribution of the input variable <math>x</math>. In practice, the methods select the optimal function by minimizing the empirical risk:


$$ R_e(g) = \sum_{i=1}^n (y_i - g(x_i))^2 $$
$$ R_e(g) = \sum_{i=1}^n (y_i - g(x_i))^2 $$


When the problem involves different tasks, the model should optimize for each data point depending on the given task. Let <math>f_j(x)</math> be the true ground-truth function for each task <math> j </math>. Therefore, for some input variable <math> x_i </math>, an ideal model <math>g</math> would predict <math> g(x_i) = f_j(x_i) </math>. With this, the risk functional can be modified to fit this new task for traditional supervised learning methods.
To minimize the risk function, the theoretically optimal solution is <math> f(x) </math>.
 
When the problem involves different tasks, the model should optimize for each data point depending on the given task. Let <math>f_j(x)</math> be the true ground-truth function for each task <math> j </math>. Therefore, for some input variable <math> x_i </math>, an ideal model <math>g</math> would predict <math> g(x_i) = f_j(x_i) </math>. With this, the risk function can be modified to fit this new task for traditional supervised learning methods.


$$ R(g) = \int_x \sum_{j=1}^n (f_j(x) - g(x))^2 p(f_j) p(x) \; \mathrm{d}x $$
$$ R(g) = \int_x \sum_{j=1}^n (f_j(x) - g(x))^2 p(f_j) p(x) \; \mathrm{d}x $$


We call <math> (f_j(x) - g(x))^2 p(f_j) </math> the '''confusing multiple mappings'''. Then the optimal solution <math>g^*(x)</math> to the mapping is <math>\bar{f}(x) = \sum_{j=1}^n p(f_j) f_j(x)</math> under this risk functional. However, the optimal solution is not conditional on the specific task at hand but rather on the entire ground-truth functions. Therefore, for every non-trivial set of tasks where <math>f_u(x) \neq f_v(x)</math> for some input <math>x</math> and <math>u \neq v</math>, <math>R(g^*) > 0</math> which implies that there is an unavoidable confusion risk.
We call <math> (f_j(x) - g(x))^2 p(f_j) </math> the '''confusing multiple mappings'''. Then the optimal solution <math>g^*(x)</math> is <math>\bar{f}(x) = \sum_{j=1}^n p(f_j) f_j(x)</math>. However, the optimal solution is not conditional on the specific task at hand but rather on the entire ground-truth functions. The solution represents a mixed probably model instead of knowing the exact tasks and their correpsonding individual probability distribution. Therefore, for every non-trivial set of tasks where <math>f_u(x) \neq f_v(x)</math> for some input <math>x</math> and <math>u \neq v</math>, <math>R(g^*) > 0</math> which implies that there is an unavoidable confusion risk.


== Learning Functions of CSL ==
== Learning Functions of CSL ==
Line 54: Line 58:
* '''Mapping function''' &mdash; mapping relation from input to the output of every learned task
* '''Mapping function''' &mdash; mapping relation from input to the output of every learned task


Suppose there are <math>n</math> ground-truth mappings <math>\{f_j : 1 \leq j \leq n\}</math> that we wish to approximate with a set of mapping functions <math>\{g_k : 1 \leq k \leq l\}</math>. The authors define the deconfusing function as an indicator function <math>h(x, y, g_k) </math> which takes some sample <math>(x,y)</math> and determines whether the sample is assigned to task <math>g_k</math>. Under the CSL framework, the risk functional (mean squared loss) is  
Suppose there are <math>n</math> ground-truth mappings <math>\{f_j : 1 \leq j \leq n\}</math> that we wish to approximate with a set of mapping functions <math>\{g_k : 1 \leq k \leq l\}</math>. The authors define the deconfusing function as an indicator function <math>h(x, y, g_k) </math> which takes some sample <math>(x,y)</math> and determines whether the sample is assigned to task <math>g_k</math>. Under the CSL framework, the risk functional (using MSE loss) is  


$$ R(g,h) = \int_x \sum_{j,k} (f_j(x) - g_k(x))^2 \; h(x, f_j(x), g_k) \;p(f_j) \; p(x) \;\mathrm{d}x $$
$$ R(g,h) = \int_x \sum_{j,k} (f_j(x) - g_k(x))^2 \; h(x, f_j(x), g_k) \;p(f_j) \; p(x) \;\mathrm{d}x $$
Line 61: Line 65:


$$R_e(g,h) = \sum_{i=1}^m \sum_{k=1}^n |y_i - g_k(x_i)|^2 \cdot h(x_i, y_i, g_k) $$
$$R_e(g,h) = \sum_{i=1}^m \sum_{k=1}^n |y_i - g_k(x_i)|^2 \cdot h(x_i, y_i, g_k) $$
The risk metric of every sample affects only its assigned task.


== Theoretical Results ==
== Theoretical Results ==
Line 73: Line 79:


''for each <math>k=1,..., n</math> that makes the expected risk function of the CSL problem zero.''
''for each <math>k=1,..., n</math> that makes the expected risk function of the CSL problem zero.''
However, necessity constraints are needed to avoid meaningless trivial solutions in all optimal risk solutions.


'''Theorem 2 (Error Bound of CSL)'''
'''Theorem 2 (Error Bound of CSL)'''
Line 81: Line 89:
''where <math>\alpha</math> is the total parameters of learning functions <math>g, h</math>, <math>B</math>  is the upper bound of one sample's risk, <math>m</math> is the size of training data and''
''where <math>\alpha</math> is the total parameters of learning functions <math>g, h</math>, <math>B</math>  is the upper bound of one sample's risk, <math>m</math> is the size of training data and''
$$\epsilon(m) = 4 \; \frac{\tau (\ln \frac{2m}{\tau} + 1) - \ln \eta / 4}{m}$$
$$\epsilon(m) = 4 \; \frac{\tau (\ln \frac{2m}{\tau} + 1) - \ln \eta / 4}{m}$$
This theorem shows the method of empirical risk minimization is valid in the CSL framework. Moreover, the assumed number of tasks affects the VC dimension of the learning functions, which is positively related to the generalization error. Therefore, to make the training risk small, we need to choose the ''minimum number'' of tasks when determining the task.


= CSL-Net =
= CSL-Net =
In this section, the authors describe how to implement and train a network for CSL.
In this section, the authors describe how to implement and train a network for CSL, including the stucture of CSL-Net and Iterative deconfusing algorithm.


== The Structure of CSL-Net ==
== The Structure of CSL-Net ==
Line 105: Line 115:
3 data sets are used to compare CSL to existing methods, 1 function regression task, and 2 image classification tasks.  
3 data sets are used to compare CSL to existing methods, 1 function regression task, and 2 image classification tasks.  


'''Function Regression''': The function regression data comes in the form of <math>(x_i,y_i),i=1,...,m</math> pairs. However, unlike typical regression problems, there are multiple <math>f_j(x),j=1,...,n</math> mapping functions, so the goal is to recover both the mapping functions <math>f_j</math> as well as determine which mapping function corresponds to each of the <math>m</math> observations. 3 scalar-valued, scalar-input functions that intersect at several points with each other have been chosen as the different tasks.  
'''Function Regression''': The function regression data comes in the form of <math>(x_i,y_i),i=1,...,m</math> pairs. However, unlike typical regression problems, there are multiple <math>f_j(x),j=1,...,n</math> mapping functions, so the goal is to reproduce both the mapping functions <math>f_j</math> as well as determine which mapping function corresponds to each of the <math>m</math> observations. 3 scalar-valued, scalar-input functions that intersect at several points with each other have been chosen as the different tasks.  


'''Colorful-MNIST''': The first image classification data set consists of the MNIST digit data that has been colored. Each observation in this modified set consists of a colored image (<math>x_i</math>) and either the color, or the digit it represents (<math>y_i</math>). The goal is to recover the classification task ("color" or "digit") for each observation and construct the 2 classifiers for both tasks.   
'''Colorful-MNIST''': The first image classification data set consists of digit data in a range of 0 to 9, each of which is in a single color among the eight different colors. Each observation in this modified set consists of a colored image (<math>x_i</math>) and a label (<math>y_i</math>) that represents either the corresponding color, or the digit. The goal is to reproduce the classification task ("color" or "digit") for each observation and construct the 2 classifiers for both tasks.   


'''Kaggle Fashion Product''': This data set has more observations than the "colored-MNIST" data and consists of pictures labeled with either the “Gender”, “Category”, and “Color” of the clothing item.
'''Kaggle Fashion Product''': The second image classification data set consists of several fashion-related objects labeled from any of the 3 criteria: “gender”, “category”, and “main color”, whose number of observations is larger than that of the "colored-MNIST" data set.


==Use of Pre-Trained CNN Feature Layers==
==Use of Pre-Trained CNN Feature Layers==


In the Kaggle Fashion Product experiment, CSL trains fully-connected layers that have been attached to feature-identifying layers from pre-trained Convolutional Neural Networks.
In the Kaggle Fashion Product experiment, CSL trains fully-connected layers that have been attached to feature-identifying layers from pre-trained Convolutional Neural Networks. The CSL methods autonomously learned three tasks which corresponded exactly to “Gender”,
“Category”, and “Color” as we see it.


==Metrics of Confusing Supervised Learning==
==Metrics of Confusing Supervised Learning==
Line 128: Line 139:


$$ \alpha_L(j) = \operatorname{max}_k\frac{1}{m}\sum_{i=1}^m 1-\dfrac{|g_k(x_i)-f_j(x_i)|}{|f_j(x_i)|}$$
$$ \alpha_L(j) = \operatorname{max}_k\frac{1}{m}\sum_{i=1}^m 1-\dfrac{|g_k(x_i)-f_j(x_i)|}{|f_j(x_i)|}$$
The purpose of this measure arises from the fact that, in addition to learning mapping allocations like humans, machines should be able to approximate all mapping functions accurately in order to provide corresponding labels. The Label Prediction Accuracy measure captures the exchange equivalence of the following task: each mapping contains its ground-truth output, and machines should be predicting the correct output that is close to the ground-truth. 


==Results==
==Results==
Line 133: Line 146:
Given confusing data, CSL performs better than traditional supervised learning methods, Pseudo-Label(Lee, 2013), and SMiLE(Tan et al., 2017). This is demonstrated by CSL's <math>\alpha_L</math> scores of around 95%, compared to <math>\alpha_L</math> scores of under 50% for the other methods. This supports the assertion that traditional methods only learn the means of all the ground-truth mapping functions when presented with confusing data.
Given confusing data, CSL performs better than traditional supervised learning methods, Pseudo-Label(Lee, 2013), and SMiLE(Tan et al., 2017). This is demonstrated by CSL's <math>\alpha_L</math> scores of around 95%, compared to <math>\alpha_L</math> scores of under 50% for the other methods. This supports the assertion that traditional methods only learn the means of all the ground-truth mapping functions when presented with confusing data.


'''Function Regression''': In order to "correctly" partition the observations into the correct tasks, a 5-shot warm-up was used. In this situation, the CSL methods work well learning the ground-truth. That means the initialization of the neural network is set up properly.
'''Function Regression''': To "correctly" partition the observations into the correct tasks, a 5-shot warm-up was used. In this situation, the CSL methods work well in learning the ground-truth. That means the initialization of the neural network is set up properly.


'''Image Classification''': Visualizations created through Spectral embedding confirm the task labeling proficiency of the deconfusing neural network <math>h</math>.
'''Image Classification''': Visualizations created through Spectral embedding confirm the task labelling proficiency of the deconfusing neural network <math>h</math>.


The classification and function prediction accuracy of CSL are comparable to supervised learning programs that have been given access to the ground-truth labels.
The classification and function prediction accuracy of CSL are comparable to supervised learning programs that have been given access to the ground-truth labels.
Line 143: Line 156:
CSL also had better accuracy than traditional supervised learning methods, Pseudo-Label(Lee, 2013), and SMiLE(Tan et al., 2017) when presented with partially labelled multi-label data <math>(x_i,y_i)</math>, where <math>y_i</math> is a <math>n</math>-long indicator vector for whether the image <math>(x_i,y_i)</math> corresponds to each of the <math>n</math> labels.
CSL also had better accuracy than traditional supervised learning methods, Pseudo-Label(Lee, 2013), and SMiLE(Tan et al., 2017) when presented with partially labelled multi-label data <math>(x_i,y_i)</math>, where <math>y_i</math> is a <math>n</math>-long indicator vector for whether the image <math>(x_i,y_i)</math> corresponds to each of the <math>n</math> labels.


Applications of multi-label classification include building a recommendation system, social media targeting, as well as detecting adverse drug reaction from text.
Applications of multi-label classification include building a recommendation system, social media targeting, as well as detecting adverse drug reactions from the text.
 
Multi-label can be used to improve the syndrome diagnosis of a patient by focusing on multiple syndromes instead of a single syndrome.


==Limitations==
==Limitations==
Line 153: Line 168:
= Conclusion =
= Conclusion =


This paper proposes the CSL method for tackling the multi-task learning problem with manual task annotations in the input data. The model obtains a basic task concept by differentiating multiple mappings. The paper also demonstrates that the CSL method is an important step to moving from Narrow AI towards General AI for multi-task learning.
This paper proposes the CSL method for tackling the multi-task learning problem without manual task annotations from basic input data. The model obtains a basic task concept by learning the minimum risk for confusing samples from differentiating multiple mappings. The paper also demonstrates that the CSL method is an important step to moving from Narrow AI towards General AI for multi-task learning.


However, there are some limitations that can be improved for future work:
However, some limitations can be improved for future work:
The repeated training process of determining the lowest best task number that has the closest to zero causes inefficiency in the learning process; The current algorithm is difficult to learn basic features directly through the CNN structure, and it is necessary to learn and train a fully connected network based on CNN features in the experiment.
 
- The repeated training process of determining the lowest best task number that has the closest to zero causes inefficiency in the learning process;  
 
- The current algorithm is difficult to learn basic features directly through a CNN structure and understand tasks simultaneously by training a full-connect network. However, this limitation does not affect the effectiveness of our algorithm in learning confusing data based on pre-trained features.


= Critique =
= Critique =
Line 163: Line 181:


Human task annotation is also imperfect, so one additional application of CSL may be to attempt to flag task annotation errors made by humans, such as in sorting comments for items sold by online retailers; concerned customers, in particular, may not correctly label their comments as "refund", "order didn't arrive", "order damaged", "how good the item is" etc.
Human task annotation is also imperfect, so one additional application of CSL may be to attempt to flag task annotation errors made by humans, such as in sorting comments for items sold by online retailers; concerned customers, in particular, may not correctly label their comments as "refund", "order didn't arrive", "order damaged", "how good the item is" etc.
Compared to the standard supervised learning, Multi-label learning can associate a training sample with multiple category tags at the same time. It can assign multiple labels to some hidden instances and can be reduced to standard supervised learning by limiting the number of class labels per instance.


This algorithm will also have a huge issue in scaling, as the proposed method requires repeated training processes, so it might be too expensive for researchers to implement and improve on this algorithm.
This algorithm will also have a huge issue in scaling, as the proposed method requires repeated training processes, so it might be too expensive for researchers to implement and improve on this algorithm.


This research paper should have included a plot on loss (of both functions) against epochs in the paper. A common issue with fixing the parameters of one network and updating the other is the variability during training. This is prevalent in other algorithms with similar training methods such as generative adversarial networks (GAN). For instance, ''mode collapse'' is the issue of one network stuck in local minima and other networks that rely on this network may receive incorrect signals during backpropagation. In the case of CSL-Net, since the Deconfusing-Net directly relies on Mapping-Net for training labels, if the Mapping-Net is unable to sufficiently converge, the Deconfusing-Net may incorrectly learn the mapping from inputs to the task. For data with high noise, oscillations may severely prolong the time needed for converge because of the strong correlation in prediction between the two networks.
This research paper should have included a plot on loss (of both functions) against epochs in the paper. A common issue with fixing the parameters of one network and updating the other is the variability during training. This is prevalent in other algorithms with similar training methods such as generative adversarial networks (GAN). For instance, ''mode collapse'' is the issue of one network stuck in local minima and other networks that rely on this network may receive incorrect signals during backpropagation. In the case of CSL-Net, since the Deconfusing-Net directly relies on Mapping-Net for training labels, if the Mapping-Net is unable to sufficiently converge, the Deconfusing-Net may incorrectly learn the mapping from inputs to the task. For data with high noise, oscillations may severely prolong the time needed to converge because of the strong correlation in prediction between the two networks.


- It would be interesting to see this implemented in more examples, to test the robustness of different types of data.
It would be interesting to see this implemented in more examples, to test the robustness of different types of data. The validation tasks chosen by data are all very simple, and CSL is actually not necessary for those tasks. For the colored MNIST data, a simple function can be written to distinguish the color label from the number label. The same problem applied to the Kaggle Fashion product dataset. The candidate label can be easily classified into different tasks by some wording analysis or meaning classification program or even manual classification. Even though the idea discussed by authors are interesting, the examples suggested by authors seem to suggest very limited or even unnecessary application. In most cases, it is more beneficial to treat the Confusing Multi-task Data problems separately into two distinct stages: we classify the tasks first according to the meaning of the label, and then we perform a multi-class/multi-label training process.


Even though this paper has already included some examples when testing the CSL in experiments, it will be better to include more detailed examples for partial-label in the "Application of Multi-label Learning" section.
When using this framework for classification, the order of the one-hot classification labels for each task will likely influence the relationships learned between each task, since the same output header is used for all tasks. This may be why this method fails to learn low-level representations and requires pretraining. I would like to see more explanation in the paper about why this isn't a problem if it was investigated.
 
It would be a good idea to include comparison details in the summary to make the results and the conclusion more convincing. For instance, though the paper introduced the result generated using confusion data, and provide some applications for multi-label learning, these two sections still fell short and could use some technical details as supporting evidence.
 
It is interesting to investigate if the order of adding tasks will influence the model performance.
 
It would be interesting to see the effectiveness of applying CSL in face recognition, such that not only does the algorithm map the face to identity, it also categorizes the face based on other features like beard/no beard and glasses/no glasses simultaneously.
 
It would be better for the researchers to compare the efficiency of this approach with other models.
 
For pattern recognition,pre-trained features were used in the algorithm. It would be interesting to see how the effectiveness of the model changes if we train it with data directly from the CNN structure in the future.
 
Basically given a confused dataset CSL finds the important tasks or labels from the dataset as can be seen from the fruit example. In the example, fruits are grouped under their names, their tastes, and their color, when CSL is given a mixed dataset. Hence given an unstructured data, unlabeled, confused dataset CSL helps in finding the labels, which in turn can help in cleaning the dataset and further in preparing high-quality training data set which is very important in different ML algorithms. Since at present preparing these dataset requires manual data annotations, CSL can save time in that process.
 
For the Colorful-Mnist data set, the goal is to understand the concept of multiple classification tasks from these examples. All inputs have multiple classification tasks. Each observed sample only represents the classification result of one task, and the task from which the sample comes is unknown.
 
It would be nice to know why the given metrics of confusing supervised learning are used. The authors should have used several different metrics and show that CSL's overall performs better than other methods. And what are "the other methods" referring to? algorithm
 
In the Iterative Deconfusing algorithm section, the Training of Mapping-Net needs more explanation. The authors should specify what it is doing before showing its equations.
 
For the results section, it would be more intuitive and stronger if the author provide more detail on these two methods and add a plot to support the claim. Based on the text, it might not be an obvious comparison.
 
It will be interesting to see if this model can work for other datasets such as CIFAR-10, CIFAR-100, and ImageNet and how well it will perform on those datasets.
 
Since this is the summary of the paper, it would be better if the introduction can concentrate more on  multi-tasking and be trimmed a little by shorten the fruit example.


= References =
= References =
Line 183: Line 227:


[5] Chavdarova, Tatjana, and François Fleuret. "Sgan: An alternative training of generative adversarial networks." In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 9407-9415. 2018.
[5] Chavdarova, Tatjana, and François Fleuret. "Sgan: An alternative training of generative adversarial networks." In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 9407-9415. 2018.
[6] Guo-Ping Liu, Jian-Jun Yan, Yi-Qin Wang, Jing-Jing Fu, Zhao-Xia Xu, Rui Guo, Peng Qian, "Application of Multilabel Learning Using the Relevant Feature for Each Label in Chronic Gastritis Syndrome Diagnosis", Evidence-Based Complementary and Alternative Medicine, vol. 2012, Article ID 135387, 9 pages, 2012. https://doi.org/10.1155/2012/135387

Latest revision as of 14:35, 7 December 2020

Presented By

Qianlin Song, William Loh, Junyue Bai, Phoebe Choi

Introduction

Narrow AI is an artificial intelligence that outperforms humans in a narrowly defined task. The application of Narrow AI is becoming more and more common. For example, Narrow AI can be used for spam filtering, music recommendation services, assisting doctors to make data-driven decisions, and even self-driving cars. One of the most famous integrated forms of Narrow AI is Apple's Siri. Siri has no self-awareness or genuine intelligence, and hence often has challenges performing tasks outside its range of abilities. However, the widespread use of Narrow AI in important infrastructure functions raises some concerns. Some people think that the characteristics of Narrow AI make it fragile, and when neural networks can be used to control important systems (such as power grids, financial transactions), alternatives may be more inclined to avoid risks. While these machines help companies improve efficiency and cut costs, the limitations of Narrow AI encouraged researchers to look into General AI.

General AI is a machine that can apply its learning to different contexts, which closely resembles human intelligence. This paper attempts to generalize the multi-task learning system that learns from data from multiple classification tasks. For an isolated and very difficult task, the artificial intelligence may not learn it very well. For instance, a net with pixel dimension 1000*1000 is less likely to identify complicated objects in real-world situations on the time basis. However, if it could be learned simultaneously, it would be better as the tasks can share what they learned. It is easier for the learner to learn together instead of in isolation, for example, shapes, landmarks, textures, orientation and so on. This is called Multitask Learning. One application is image recognition. In figure 1, an image of an apple corresponds to 3 labels: “red”, “apple” and “sweet”. These labels correspond to 3 different classification tasks: color, fruit, and taste.

Currently, multi-task machines require researchers to construct a task definition. Otherwise, it will end up with different outputs with the same input values. Researchers manually assign tasks to each input in the sample to train the machine. See figure 1(a). This method incurs high annotation costs and restricts the machine’s ability to mirror the human recognition process. This paper is interested in developing an algorithm that understands task concepts and performs multi-task learning without manual task annotations.

This paper proposed a new learning method called confusing supervised learning (CSL) which includes two functions: de-confusing function and mapping function. The de-confusing function allocates samples to respective tasks and the mapping function presents the relation from the input to its label within the allocated tasks. See figure 1(b). To implement the CSL, we use a risk functional to balance the effects of the de-confusing function and mapping function.

However, simply combining the two functions or networks to a single architecture is impossible, since the one-hot constraint of the outputs for the de-confusing network makes the gradient back-propagation unfeasible. This difficulty is solved by alternatively performing training for the de-confusing net and mapping net optimization in the proposed architecture CLS-Net.

Experiments for function regression and image recognition problems were constructed and compared with multi-task learning with complete information to test CSL-Net’s performance. Experiment results show that CSL-Net can learn multiple mappings for every task simultaneously and achieve the same cognition result as the current multi-task machine assigned with complete information.

Related Work

Latent variable learning

Latent variable learning aims to estimate the true function with mixed probability models. See figure 2a. In the multi-task learning problem without task annotations, we know that samples are generated from multiple distinct distributions instead of one distribution, combining multiple probability models. Thus, the latent variable learning can not fully distinguish labels into different tasks and distributions, and it is insufficient to classify the multi-task confusing samples.

Multi-task learning

Multi-task learning aims to learn multiple tasks simultaneously using a shared feature representation. By exploiting similarities and differences between tasks, the learning from one task can improve the learning of another task. (Caruana, 1997) This results in improving the overall learning efficiency since the labels in different tasks are often correlated: improving the classification result for one class also helps with other classification tasks. In multi-task learning, the input-output mapping of every task can be represented by a unified function. However, these task definitions are manually constructed, and machines need manual task annotations to learn. If such manual task annotation is absent, then the algorithm can not be performed.

Multi-label learning

Multi-label learning aims to assign an input to a set of classes/labels. See figure 2b. It is a generalization of multi-class classification, which classifies an input into one class. In multi-label learning, an input can be classified into more than one class. Unlike multi-task learning, multi-label does not consider the relationship between different label judgments, and it is assumed that each judgment is independent. An example where multi-label learning is applicable is the scenario where a website wants to automatically assign applicable tags/categories to an article. Since an article can be related to multiple categories (e.g., an article can be tagged under the politics and business categories), multi-label learning is of primary concern.

Confusing Supervised Learning

Description of the Problem

Confusing supervised learning (CSL) offers a solution to the issue at hand. A major area of improvement can be seen in the choice of risk measure. In traditional supervised learning, let [math]\displaystyle{ (x,y) }[/math] be the training samples from [math]\displaystyle{ y=f(x) }[/math], which is an identical but unknown mapping relationship. Assuming the risk measure is mean squared error (MSE), the expected risk function is

$$ R(g) = \int_x (f(x) - g(x))^2 p(x) \; \mathrm{d}x $$

where [math]\displaystyle{ p(x) }[/math] is the data distribution of the input variable [math]\displaystyle{ x }[/math]. In practice, the methods select the optimal function by minimizing the empirical risk:

$$ R_e(g) = \sum_{i=1}^n (y_i - g(x_i))^2 $$

To minimize the risk function, the theoretically optimal solution is [math]\displaystyle{ f(x) }[/math].

When the problem involves different tasks, the model should optimize for each data point depending on the given task. Let [math]\displaystyle{ f_j(x) }[/math] be the true ground-truth function for each task [math]\displaystyle{ j }[/math]. Therefore, for some input variable [math]\displaystyle{ x_i }[/math], an ideal model [math]\displaystyle{ g }[/math] would predict [math]\displaystyle{ g(x_i) = f_j(x_i) }[/math]. With this, the risk function can be modified to fit this new task for traditional supervised learning methods.

$$ R(g) = \int_x \sum_{j=1}^n (f_j(x) - g(x))^2 p(f_j) p(x) \; \mathrm{d}x $$

We call [math]\displaystyle{ (f_j(x) - g(x))^2 p(f_j) }[/math] the confusing multiple mappings. Then the optimal solution [math]\displaystyle{ g^*(x) }[/math] is [math]\displaystyle{ \bar{f}(x) = \sum_{j=1}^n p(f_j) f_j(x) }[/math]. However, the optimal solution is not conditional on the specific task at hand but rather on the entire ground-truth functions. The solution represents a mixed probably model instead of knowing the exact tasks and their correpsonding individual probability distribution. Therefore, for every non-trivial set of tasks where [math]\displaystyle{ f_u(x) \neq f_v(x) }[/math] for some input [math]\displaystyle{ x }[/math] and [math]\displaystyle{ u \neq v }[/math], [math]\displaystyle{ R(g^*) \gt 0 }[/math] which implies that there is an unavoidable confusion risk.

Learning Functions of CSL

To overcome this issue, the authors introduce two types of learning functions:

  • Deconfusing function — allocation of which samples come from the same task
  • Mapping function — mapping relation from input to the output of every learned task

Suppose there are [math]\displaystyle{ n }[/math] ground-truth mappings [math]\displaystyle{ \{f_j : 1 \leq j \leq n\} }[/math] that we wish to approximate with a set of mapping functions [math]\displaystyle{ \{g_k : 1 \leq k \leq l\} }[/math]. The authors define the deconfusing function as an indicator function [math]\displaystyle{ h(x, y, g_k) }[/math] which takes some sample [math]\displaystyle{ (x,y) }[/math] and determines whether the sample is assigned to task [math]\displaystyle{ g_k }[/math]. Under the CSL framework, the risk functional (using MSE loss) is

$$ R(g,h) = \int_x \sum_{j,k} (f_j(x) - g_k(x))^2 \; h(x, f_j(x), g_k) \;p(f_j) \; p(x) \;\mathrm{d}x $$

which can be estimated empirically with

$$R_e(g,h) = \sum_{i=1}^m \sum_{k=1}^n |y_i - g_k(x_i)|^2 \cdot h(x_i, y_i, g_k) $$

The risk metric of every sample affects only its assigned task.

Theoretical Results

This novel framework yields some theoretical results to show the viability of its construction.

Theorem 1 (Existence of Solution) With the confusing supervised learning framework, there is an optimal solution $$h^*(x, f_j(x), g_k) = \mathbb{I}[j=k]$$

$$g_k^*(x) = f_k(x)$$

for each [math]\displaystyle{ k=1,..., n }[/math] that makes the expected risk function of the CSL problem zero.

However, necessity constraints are needed to avoid meaningless trivial solutions in all optimal risk solutions.

Theorem 2 (Error Bound of CSL) With probability at least [math]\displaystyle{ 1 - \eta }[/math] simultaneously with finite VC dimension [math]\displaystyle{ \tau }[/math] of CSL learning framework, the risk measure is bounded by

$$R(\alpha) \leq R_e(\alpha) + \frac{B\epsilon(m)}{2} \left(1 + \sqrt{1 + \frac{4R_e(\alpha)}{B\epsilon(m)}}\right)$$

where [math]\displaystyle{ \alpha }[/math] is the total parameters of learning functions [math]\displaystyle{ g, h }[/math], [math]\displaystyle{ B }[/math] is the upper bound of one sample's risk, [math]\displaystyle{ m }[/math] is the size of training data and $$\epsilon(m) = 4 \; \frac{\tau (\ln \frac{2m}{\tau} + 1) - \ln \eta / 4}{m}$$

This theorem shows the method of empirical risk minimization is valid in the CSL framework. Moreover, the assumed number of tasks affects the VC dimension of the learning functions, which is positively related to the generalization error. Therefore, to make the training risk small, we need to choose the minimum number of tasks when determining the task.

CSL-Net

In this section, the authors describe how to implement and train a network for CSL, including the stucture of CSL-Net and Iterative deconfusing algorithm.

The Structure of CSL-Net

Two neural networks, deconfusing-net and mapping-net are trained to implement two learning function variables in empirical risk. The optimization target of the training algorithm is: $$\min_{g, h} R_e = \sum_{i=1}^{m}\sum_{k=1}^{n} (y_i - g_k(x_i))^2 \cdot h(x_i, y_i; g_k)$$

The mapping-net is corresponding to functions set [math]\displaystyle{ g_k }[/math], where [math]\displaystyle{ y_k = g_k(x) }[/math] represents the output of one certain task. The deconfusing-net is corresponding to function h, whose input is a sample [math]\displaystyle{ (x,y) }[/math] and output is an n-dimensional one-hot vector. This output vector determines which task the sample [math]\displaystyle{ (x,y) }[/math] should be assigned to. The core difficulty of this algorithm is that the risk function cannot be optimized by gradient back-propagation due to the constraint of one-hot output from deconfusing-net. Approximation of softmax will lead the deconfusing-net output into a non-one-hot form, which results in meaningless trivial solutions.

Iterative Deconfusing Algorithm

To overcome the training difficulty, the authors divide the empirical risk minimization into two local optimization problems. In each single-network optimization step, the parameters of one network are updated while the parameters of another remain fixed. With one network's parameters unchanged, the problem can be solved by a gradient descent method of neural networks.

Training of Mapping-Net: With function h from deconfusing-net being determined, the goal is to train every mapping function [math]\displaystyle{ g_k }[/math] with its corresponding sample [math]\displaystyle{ (x_i^k, y_i^k) }[/math]. The optimization problem becomes: [math]\displaystyle{ \displaystyle \min_{g_k} L_{map}(g_k) = \sum_{i=1}^{m_k} \mid y_i^k - g_k(x_i^k)\mid^2 }[/math]. Back-propagation algorithm can be applied to solve this optimization problem.

Training of Deconfusing-Net: The task allocation is re-evaluated during the training phase while the parameters of the mapping-net remain fixed. To minimize the original risk, every sample [math]\displaystyle{ (x, y) }[/math] will be assigned to [math]\displaystyle{ g_k }[/math] that is closest to label y among all different [math]\displaystyle{ k }[/math]s. Mapping-net thus provides a temporary solution for deconfusing-net: [math]\displaystyle{ \hat{h}(x_i, y_i) = arg \displaystyle\min_{k} \mid y_i - g_k(x_i)\mid^2 }[/math]. The optimization becomes: [math]\displaystyle{ \displaystyle \min_{h} L_{dec}(h) = \sum_{i=1}^{m} \mid {h}(x_i, y_i) - \hat{h}(x_i, y_i)\mid^2 }[/math]. Similarly, the optimization problem can be solved by updating the deconfusing-net with a back-propagation algorithm.

The two optimization stages are carried out alternately until the solution converges.

Experiment

Setup

3 data sets are used to compare CSL to existing methods, 1 function regression task, and 2 image classification tasks.

Function Regression: The function regression data comes in the form of [math]\displaystyle{ (x_i,y_i),i=1,...,m }[/math] pairs. However, unlike typical regression problems, there are multiple [math]\displaystyle{ f_j(x),j=1,...,n }[/math] mapping functions, so the goal is to reproduce both the mapping functions [math]\displaystyle{ f_j }[/math] as well as determine which mapping function corresponds to each of the [math]\displaystyle{ m }[/math] observations. 3 scalar-valued, scalar-input functions that intersect at several points with each other have been chosen as the different tasks.

Colorful-MNIST: The first image classification data set consists of digit data in a range of 0 to 9, each of which is in a single color among the eight different colors. Each observation in this modified set consists of a colored image ([math]\displaystyle{ x_i }[/math]) and a label ([math]\displaystyle{ y_i }[/math]) that represents either the corresponding color, or the digit. The goal is to reproduce the classification task ("color" or "digit") for each observation and construct the 2 classifiers for both tasks.

Kaggle Fashion Product: The second image classification data set consists of several fashion-related objects labeled from any of the 3 criteria: “gender”, “category”, and “main color”, whose number of observations is larger than that of the "colored-MNIST" data set.

Use of Pre-Trained CNN Feature Layers

In the Kaggle Fashion Product experiment, CSL trains fully-connected layers that have been attached to feature-identifying layers from pre-trained Convolutional Neural Networks. The CSL methods autonomously learned three tasks which corresponded exactly to “Gender”, “Category”, and “Color” as we see it.

Metrics of Confusing Supervised Learning

There are two measures of accuracy used to evaluate and compare CSL to other methods, corresponding respectively to the accuracy of the task labeling and the accuracy of the learned mapping function.

Task Prediction Accuracy: [math]\displaystyle{ \alpha_T(j) }[/math] is the average number of times the learned deconfusing function [math]\displaystyle{ h }[/math] agrees with the task-assignment ability of humans [math]\displaystyle{ \tilde h }[/math] on whether each observation in the data "is" or "is not" in task [math]\displaystyle{ j }[/math].

$$ \alpha_T(j) = \operatorname{max}_k\frac{1}{m}\sum_{i=1}^m I[h(x_i,y_i;f_k),\tilde h(x_i,y_i;f_j)]$$

The max over [math]\displaystyle{ k }[/math] is taken because we need to determine which learned task corresponds to which ground-truth task.

Label Prediction Accuracy: [math]\displaystyle{ \alpha_L(j) }[/math] again chooses [math]\displaystyle{ f_k }[/math], the learned mapping function that is closest to the ground-truth of task [math]\displaystyle{ j }[/math], and measures its average absolute accuracy compared to the ground-truth of task [math]\displaystyle{ j }[/math], [math]\displaystyle{ f_j }[/math], across all [math]\displaystyle{ m }[/math] observations.

$$ \alpha_L(j) = \operatorname{max}_k\frac{1}{m}\sum_{i=1}^m 1-\dfrac{|g_k(x_i)-f_j(x_i)|}{|f_j(x_i)|}$$

The purpose of this measure arises from the fact that, in addition to learning mapping allocations like humans, machines should be able to approximate all mapping functions accurately in order to provide corresponding labels. The Label Prediction Accuracy measure captures the exchange equivalence of the following task: each mapping contains its ground-truth output, and machines should be predicting the correct output that is close to the ground-truth.

Results

Given confusing data, CSL performs better than traditional supervised learning methods, Pseudo-Label(Lee, 2013), and SMiLE(Tan et al., 2017). This is demonstrated by CSL's [math]\displaystyle{ \alpha_L }[/math] scores of around 95%, compared to [math]\displaystyle{ \alpha_L }[/math] scores of under 50% for the other methods. This supports the assertion that traditional methods only learn the means of all the ground-truth mapping functions when presented with confusing data.

Function Regression: To "correctly" partition the observations into the correct tasks, a 5-shot warm-up was used. In this situation, the CSL methods work well in learning the ground-truth. That means the initialization of the neural network is set up properly.

Image Classification: Visualizations created through Spectral embedding confirm the task labelling proficiency of the deconfusing neural network [math]\displaystyle{ h }[/math].

The classification and function prediction accuracy of CSL are comparable to supervised learning programs that have been given access to the ground-truth labels.

Application of Multi-label Learning

CSL also had better accuracy than traditional supervised learning methods, Pseudo-Label(Lee, 2013), and SMiLE(Tan et al., 2017) when presented with partially labelled multi-label data [math]\displaystyle{ (x_i,y_i) }[/math], where [math]\displaystyle{ y_i }[/math] is a [math]\displaystyle{ n }[/math]-long indicator vector for whether the image [math]\displaystyle{ (x_i,y_i) }[/math] corresponds to each of the [math]\displaystyle{ n }[/math] labels.

Applications of multi-label classification include building a recommendation system, social media targeting, as well as detecting adverse drug reactions from the text.

Multi-label can be used to improve the syndrome diagnosis of a patient by focusing on multiple syndromes instead of a single syndrome.

Limitations

Number of Tasks: The number of tasks is determined by increasing the task numbers progressively and testing the performance. Ideally, a better way of deciding the number of tasks is expected rather than increasing it one by one and seeing which is the minimum number of tasks that gives the smallest risk. Adding low-quality constraints to deconfusing-net is a reasonable solution to this problem.

Learning of Basic Features: The CSL framework is not good at learning features. So far, a pre-trained CNN backbone is needed for complicated image classification problems. Even though the effectiveness of the proposed algorithm in learning confusing data based on pre-trained features hasn't been affected, the full-connect network can only be trained based on learned CNN features. It is still a challenge for the current algorithm to learn basic features directly through a CNN structure and understand tasks simultaneously.

Conclusion

This paper proposes the CSL method for tackling the multi-task learning problem without manual task annotations from basic input data. The model obtains a basic task concept by learning the minimum risk for confusing samples from differentiating multiple mappings. The paper also demonstrates that the CSL method is an important step to moving from Narrow AI towards General AI for multi-task learning.

However, some limitations can be improved for future work:

- The repeated training process of determining the lowest best task number that has the closest to zero causes inefficiency in the learning process;

- The current algorithm is difficult to learn basic features directly through a CNN structure and understand tasks simultaneously by training a full-connect network. However, this limitation does not affect the effectiveness of our algorithm in learning confusing data based on pre-trained features.

Critique

The classification accuracy of CSL was made with algorithms not designed to deal with confusing data and which do not first classify the task of each observation.

Human task annotation is also imperfect, so one additional application of CSL may be to attempt to flag task annotation errors made by humans, such as in sorting comments for items sold by online retailers; concerned customers, in particular, may not correctly label their comments as "refund", "order didn't arrive", "order damaged", "how good the item is" etc.

Compared to the standard supervised learning, Multi-label learning can associate a training sample with multiple category tags at the same time. It can assign multiple labels to some hidden instances and can be reduced to standard supervised learning by limiting the number of class labels per instance.

This algorithm will also have a huge issue in scaling, as the proposed method requires repeated training processes, so it might be too expensive for researchers to implement and improve on this algorithm.

This research paper should have included a plot on loss (of both functions) against epochs in the paper. A common issue with fixing the parameters of one network and updating the other is the variability during training. This is prevalent in other algorithms with similar training methods such as generative adversarial networks (GAN). For instance, mode collapse is the issue of one network stuck in local minima and other networks that rely on this network may receive incorrect signals during backpropagation. In the case of CSL-Net, since the Deconfusing-Net directly relies on Mapping-Net for training labels, if the Mapping-Net is unable to sufficiently converge, the Deconfusing-Net may incorrectly learn the mapping from inputs to the task. For data with high noise, oscillations may severely prolong the time needed to converge because of the strong correlation in prediction between the two networks.

It would be interesting to see this implemented in more examples, to test the robustness of different types of data. The validation tasks chosen by data are all very simple, and CSL is actually not necessary for those tasks. For the colored MNIST data, a simple function can be written to distinguish the color label from the number label. The same problem applied to the Kaggle Fashion product dataset. The candidate label can be easily classified into different tasks by some wording analysis or meaning classification program or even manual classification. Even though the idea discussed by authors are interesting, the examples suggested by authors seem to suggest very limited or even unnecessary application. In most cases, it is more beneficial to treat the Confusing Multi-task Data problems separately into two distinct stages: we classify the tasks first according to the meaning of the label, and then we perform a multi-class/multi-label training process.

When using this framework for classification, the order of the one-hot classification labels for each task will likely influence the relationships learned between each task, since the same output header is used for all tasks. This may be why this method fails to learn low-level representations and requires pretraining. I would like to see more explanation in the paper about why this isn't a problem if it was investigated.

It would be a good idea to include comparison details in the summary to make the results and the conclusion more convincing. For instance, though the paper introduced the result generated using confusion data, and provide some applications for multi-label learning, these two sections still fell short and could use some technical details as supporting evidence.

It is interesting to investigate if the order of adding tasks will influence the model performance.

It would be interesting to see the effectiveness of applying CSL in face recognition, such that not only does the algorithm map the face to identity, it also categorizes the face based on other features like beard/no beard and glasses/no glasses simultaneously.

It would be better for the researchers to compare the efficiency of this approach with other models.

For pattern recognition,pre-trained features were used in the algorithm. It would be interesting to see how the effectiveness of the model changes if we train it with data directly from the CNN structure in the future.

Basically given a confused dataset CSL finds the important tasks or labels from the dataset as can be seen from the fruit example. In the example, fruits are grouped under their names, their tastes, and their color, when CSL is given a mixed dataset. Hence given an unstructured data, unlabeled, confused dataset CSL helps in finding the labels, which in turn can help in cleaning the dataset and further in preparing high-quality training data set which is very important in different ML algorithms. Since at present preparing these dataset requires manual data annotations, CSL can save time in that process.

For the Colorful-Mnist data set, the goal is to understand the concept of multiple classification tasks from these examples. All inputs have multiple classification tasks. Each observed sample only represents the classification result of one task, and the task from which the sample comes is unknown.

It would be nice to know why the given metrics of confusing supervised learning are used. The authors should have used several different metrics and show that CSL's overall performs better than other methods. And what are "the other methods" referring to? algorithm

In the Iterative Deconfusing algorithm section, the Training of Mapping-Net needs more explanation. The authors should specify what it is doing before showing its equations.

For the results section, it would be more intuitive and stronger if the author provide more detail on these two methods and add a plot to support the claim. Based on the text, it might not be an obvious comparison.

It will be interesting to see if this model can work for other datasets such as CIFAR-10, CIFAR-100, and ImageNet and how well it will perform on those datasets.

Since this is the summary of the paper, it would be better if the introduction can concentrate more on multi-tasking and be trimmed a little by shorten the fruit example.

References

[1] Su, Xin, et al. "Task Understanding from Confusing Multi-task Data."

[2] Caruana, R. (1997) "Multi-task learning"

[3] Lee, D.-H. Pseudo-label: The simple and efficient semi-supervised learning method for deep neural networks. Workshop on challenges in representation learning, ICML, vol. 3, 2013, pp. 2–8.

[4] Tan, Q., Yu, Y., Yu, G., and Wang, J. Semi-supervised multi-label classification using incomplete label information. Neurocomputing, vol. 260, 2017, pp. 192–202.

[5] Chavdarova, Tatjana, and François Fleuret. "Sgan: An alternative training of generative adversarial networks." In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 9407-9415. 2018.

[6] Guo-Ping Liu, Jian-Jun Yan, Yi-Qin Wang, Jing-Jing Fu, Zhao-Xia Xu, Rui Guo, Peng Qian, "Application of Multilabel Learning Using the Relevant Feature for Each Label in Chronic Gastritis Syndrome Diagnosis", Evidence-Based Complementary and Alternative Medicine, vol. 2012, Article ID 135387, 9 pages, 2012. https://doi.org/10.1155/2012/135387