Caleb Chang

Publications

1. 

   Learning Probabilistic Responsibility Allocations for Multi-Agent Interactions

   Remy, I., Chang, C., and Leung, K.

   (preprint)

   Abstract Arxiv Bib

   Human behavior in interactive settings is shaped not only by individual objectives but also by shared constraints with others, such as safety. Understanding how people allocate responsibility, i.e., how much one deviates from their desired policy to accommodate others, can inform the design of socially compliant and trustworthy autonomous systems. In this work, we introduce a method for learning a probabilistic responsibility allocation model that captures the multimodal uncertainty inherent in multi-agent interactions. Specifically, our approach leverages the latent space of a conditional variational autoencoder, combined with techniques from multi-agent trajectory forecasting, to learn a distribution over responsibility allocations conditioned on scene and agent context. Although ground-truth responsibility labels are unavailable, the model remains tractable by incorporating a differentiable optimization layer that maps responsibility allocations to induced controls, which are available. We evaluate our method on the INTERACTION driving dataset and demonstrate that it not only achieves strong predictive performance but also provides interpretable insights, through the lens of responsibility, into patterns of multi-agent interaction.

   @article{RemyChangEtAl2026,
     author = {Remy, I. and Chang, C. and Leung, K.},
     journal = {{{Available at }\url{https://arxiv.org/abs/2604.13128}}},
     title = {{Learning Probabilistic Responsibility Allocations for Multi-Agent Interactions}},
     year = {2026},
     arxiv = {2604.13128},
     category = {interaction},
     img = {RemyChangEtAl2026.png},
     note = {(preprint)},
     keywords = {preprint},
     owner = {karenl7}
   }

2. 

   Do as the Romans Do: Learning Universal Behaviors from Heterogeneous Agents

   Chang, C., Win Kyi, D., Jaques, N., and Leung, K.

   2026

   Abstract Arxiv Bib

   Humans often acquire new skills by observing others, since observed behaviors implicitly reveal how to act in an environment. However, observations drawn from a heterogeneous population introduce conflicting behavioral signals, making it difficult to determine which behaviors are worth imitating. We address this challenge with General Reward Inference and Disentanglement (GRID), a social learning method that extracts universally useful behaviors from a heterogeneous population of demonstrators pursuing different goals. GRID decomposes per-agent reward functions into a general reward, capturing behaviors shared across all agents, and specific rewards, capturing individual preferences and objectives. Training exclusively on the general reward provides a new paradigm of generalist pretraining. It yields a generalist agent that internalizes universal environmental competencies, such as safety and basic task proficiency, without the mode-averaging bias that afflicts standard learning from demonstration techniques. This generalist serves as a superior prior for fine-tuning to downstream tasks, including preferences unseen during training. Experiments across a synthetic basis function decomposition, multi-agent Craftax, and a continuous autonomous driving simulator (Highway-Env) confirm that GRID successfully disentangles reward structure in a semantically meaningful way, outperforms standard learning from demonstration baselines, and enables more efficient and stable specialization.

   @inproceedings{ChangWinKyiEtAl2026,
     author = {Chang, C. and Win Kyi, D. and Jaques, N. and Leung, K.},
     title = {{Do as the Romans Do: Learning Universal Behaviors from Heterogeneous Agents}},
     year = {2026},
     arxiv = {2606.18537},
     img = {ChangWinKyiEtAl2026.png},
     note = {(submitted)},
     owner = {karenl7}
   }