Task Understanding from Confusing Multi-task Data: Difference between revisions
No edit summary |
|||
| Line 8: | Line 8: | ||
= Confusing Supervised Learning = | = Confusing Supervised Learning = | ||
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, 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>p(x)</math> is the prior distribution of the input variable <math>x</math>. In practice, model optimizations are performed using the empirical risk | |||
$$ R_e(g) = \sum_{i=1}^n (y_i - g(x_i))^2 $$ | |||
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 target function for each task <math> j </math>. | |||
= CSL-Net = | = CSL-Net = | ||
Revision as of 19:05, 10 November 2020
Presented By
Qianlin Song, William Loh, Junyue Bai, Phoebe Choi
Introduction
Related Work
Confusing Supervised Learning
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, 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 prior distribution of the input variable [math]\displaystyle{ x }[/math]. In practice, model optimizations are performed using the empirical risk
$$ R_e(g) = \sum_{i=1}^n (y_i - g(x_i))^2 $$
When the problem involves different tasks, the model should optimize for each data point depending on the given task. Let [math]\displaystyle{ f_j(x) }[/math] be the target function for each task [math]\displaystyle{ j }[/math].
CSL-Net
Experiment
Conclusion
Critique
References
Su, Xin, et al. "Task Understanding from Confusing Multi-task Data."