Hierarchically Robust Representation Learning

by   Qi Qian, et al.
Alibaba Group
University of Washington

With the tremendous success of deep learning in visual tasks, the representations extracted from intermediate layers of learned models, that is, deep features, attract much attention of researchers. The previous analysis shows that those features include appropriate semantic information. By training the deep models on a large-scale benchmark data set (e.g., ImageNet), the features can work well on other tasks. In this work, we investigate this phenomenon and demonstrate that deep features can fail due to the fact that they are learned by minimizing empirical risk. When the distribution of data is different from that of the benchmark data set, the performance of deep features can degrade. Hence, we propose a hierarchically robust optimization to learn more generic features. Considering the example-level and concept-level robustness simultaneously, we formulate the problem as a distributionally robust optimization problem with Wasserstein ambiguity set constraints. An efficient algorithm with the conventional training pipeline is proposed. Experiments on benchmark data sets confirm our claim and demonstrate the effectiveness of the robust deep representations.


page 5

page 7


Learning Representations Robust to Group Shifts and Adversarial Examples

Despite the high performance achieved by deep neural networks on various...

Examining and Combating Spurious Features under Distribution Shift

A central goal of machine learning is to learn robust representations th...

Towards Universal Backward-Compatible Representation Learning

Conventional model upgrades for visual search systems require offline re...

General risk measures for robust machine learning

A wide array of machine learning problems are formulated as the minimiza...

Concept Generalization in Visual Representation Learning

Measuring concept generalization, i.e., the extent to which models train...

Pedestrian Detection aided by Deep Learning Semantic Tasks

Deep learning methods have achieved great success in pedestrian detectio...

Deep Morphological Hit-or-Miss Transform Neural Network

Neural networks have demonstrated breakthrough results in numerous appli...

Please sign up or login with your details

Forgot password? Click here to reset