This shows you the differences between two versions of the page.
— |
papers:zhang-2022 [2023/08/29 06:05] (current) leonb created |
||
---|---|---|---|
Line 1: | Line 1: | ||
+ | ===== Rich feature construction for the optimization-generalization dilemma ===== | ||
+ | |||
+ | // | ||
+ | generalization. For instance, many OoD methods | ||
+ | rely on penalty terms whose optimization is challenging. They are either too strong to optimize | ||
+ | reliably or too weak to achieve their goals. | ||
+ | In order to escape this dilemma, we propose to | ||
+ | first construct a rich representation (RFC) containing a palette of potentially useful features, ready | ||
+ | to be used by even simple models. On the one | ||
+ | hand, a rich representation provides a good initialization for the optimizer. On the other hand, | ||
+ | it also provides an inductive bias that helps OoD | ||
+ | generalization. RFC is constructed in a succession of training episodes. During each step of | ||
+ | the discovery phase, we craft a multi-objective | ||
+ | optimization criterion and its associated datasets | ||
+ | in a manner that prevents the network from using | ||
+ | the features constructed in the previous iterations. | ||
+ | During the synthesis phase, we use knowledge | ||
+ | distillation to force the network to simultaneously | ||
+ | develop all the features identified during the discovery phase. | ||
+ | RFC consistently helps six OoD methods achieve | ||
+ | top performance on challenging invariant training benchmarks, ColoredMNIST (Arjovsky et al., | ||
+ | 2020). Furthermore, | ||
+ | task, our method helps both OoD and ERM | ||
+ | methods outperform earlier compatable results | ||
+ | by at least 5%, reduce standard deviation by at | ||
+ | least 4.1%, and makes hyperparameter tuning and | ||
+ | model selection more reliable. | ||
+ | |||
+ | {{ rfc.png?500 }} | ||
+ | |||
+ | <box 99% orange> | ||
+ | Jianyu Zhang, David Lopez-Paz and Léon Bottou: | ||
+ | |||
+ | [[http:// | ||
+ | [[http:// | ||
+ | [[http:// | ||
+ | </ | ||
+ | |||
+ | @inproceedings{zhang-2022, | ||
+ | title = {Rich feature construction for the optimization-generalization dilemma}, | ||
+ | author = {Zhang, Jianyu and Lopez-Paz, David and Bottou, L{\' | ||
+ | booktitle = {International Conference on Machine Learning}, | ||
+ | pages = {26397--26411}, | ||
+ | year = {2022}, | ||
+ | organization = {PMLR}, | ||
+ | url = {http:// | ||
+ | |||