Generalization to out-of-distribution (OOD) data is a capability natural to humans yet
challenging for machines to reproduce. This is because most learning algorithms strongly …

Research on intelligent wireless network aims at the development of a human society which
is ubiquitous and mobile, simultaneously providing solutions to the coverage, capacity, and …

In this work we identify the dormant neuron phenomenon in deep reinforcement learning,
where an agent's network suffers from an increasing number of inactive neurons, thereby …

The ability to generalize well is one of the primary desiderata for models of natural language
processing (NLP), but what 'good generalization'entails and how it should be evaluated is …

Autonomous driving systems have witnessed significant development during the past years
thanks to the advance in machine learning-enabled sensing and decision-making …

Training generally-capable agents with reinforcement learning (RL) remains a significant
challenge. A promising avenue for improving the robustness of RL agents is through the use …

Foundation models have shown impressive adaptation and scalability in supervised and self-
supervised learning problems, but so far these successes have not fully translated to …

Foundation models have shown impressive adaptation and scalability in supervised and self-
supervised learning problems, but so far these successes have not fully translated to …

Many problems in RL, such as meta-RL, robust RL, generalization in RL, and temporal credit
assignment, can be cast as POMDPs. In theory, simply augmenting model-free RL with …

We study goal misgeneralization, a type of out-of-distribution robustness failure in
reinforcement learning (RL). Goal misgeneralization occurs when an RL agent retains its …