Conditional mutual information for disentangled representations in reinforcement learning
Reinforcement Learning (RL) environments can produce training data with spurious
correlations between features due to the amount of training data or its limited feature …
correlations between features due to the amount of training data or its limited feature …
10 Years of Fair Representations: Challenges and Opportunities
Fair Representation Learning (FRL) is a broad set of techniques, mostly based on neural
networks, that seeks to learn new representations of data in which sensitive or undesired …
networks, that seeks to learn new representations of data in which sensitive or undesired …
Information Bottleneck Analysis of Deep Neural Networks via Lossy Compression
I Butakov, A Tolmachev, S Malanchuk… - arXiv preprint arXiv …, 2023 - arxiv.org
The Information Bottleneck (IB) principle offers an information-theoretic framework for
analyzing the training process of deep neural networks (DNNs). Its essence lies in tracking …
analyzing the training process of deep neural networks (DNNs). Its essence lies in tracking …