Log In Sign Up

The Conditional Entropy Bottleneck

by   Ian Fischer, et al.

Much of the field of Machine Learning exhibits a prominent set of failure modes, including vulnerability to adversarial examples, poor out-of-distribution (OoD) detection, miscalibration, and willingness to memorize random labelings of datasets. We characterize these as failures of robust generalization, which extends the traditional measure of generalization as accuracy or related metrics on a held-out set. We hypothesize that these failures to robustly generalize are due to the learning systems retaining too much information about the training data. To test this hypothesis, we propose the Minimum Necessary Information (MNI) criterion for evaluating the quality of a model. In order to train models that perform well with respect to the MNI criterion, we present a new objective function, the Conditional Entropy Bottleneck (CEB), which is closely related to the Information Bottleneck (IB). We experimentally test our hypothesis by comparing the performance of CEB models with deterministic models and Variational Information Bottleneck (VIB) models on a variety of different datasets and robustness challenges. We find strong empirical evidence supporting our hypothesis that MNI models improve on these problems of robust generalization.


page 1

page 2

page 3

page 4


Revisiting Hilbert-Schmidt Information Bottleneck for Adversarial Robustness

We investigate the HSIC (Hilbert-Schmidt independence criterion) bottlen...

Deep Deterministic Information Bottleneck with Matrix-based Entropy Functional

We introduce the matrix-based Renyi's α-order entropy functional to para...

Entropy Penalty: Towards Generalization Beyond the IID Assumption

It has been shown that instead of learning actual object features, deep ...

Improving the Adversarial Robustness of NLP Models by Information Bottleneck

Existing studies have demonstrated that adversarial examples can be dire...

Gated Information Bottleneck for Generalization in Sequential Environments

Deep neural networks suffer from poor generalization to unseen environme...

Bottlenecks CLUB: Unifying Information-Theoretic Trade-offs Among Complexity, Leakage, and Utility

Bottleneck problems are an important class of optimization problems that...

The Variational Deficiency Bottleneck

We introduce a bottleneck method for learning data representations based...