publications

2026

1. 

   Effectiveness of curriculum learning depends on reward sparsity and competing optima: analysis of a tractable model of policy learning

   John J. Vastola, Ann Huang, Satpreet H. Singh, and 2 more authors

   Under review, 2026

2025

1. 

   Measuring and Controlling Solution Degeneracy across Task-Trained Recurrent Neural Networks

   Ann Huang, Satpreet H. Singh, Flavio Martinelli, and 1 more author

   In Advances in Neural Information Processing Systems, 2025

2. 

   InputDSA: Demixing, then comparing recurrent and externally driven dynamics

   Ann Huang^*, Mitchell Ostrow^*, Satpreet H. Singh, and 3 more authors

   ICLR2026, 2025

3. 

   Action chunking as conditional policy compression

   Lucy Lai^*, Ann Huang^*, and Samuel J. Gershman

   Cognition, 2025

   Abs DOI

   Many skills in our everyday lives are learned by sequencing actions towards a desired goal. The action sequence can become a “chunk” when individual actions are grouped together and executed as one unit, making them more efficient to store and execute. While chunking has been studied extensively across various domains, a puzzle remains as to why and under what conditions action chunking occurs. To tackle these questions, we develop a model of conditional policy compression—the reduction in cognitive cost by conditioning on an additional source of information—to explain the origin of chunking. We argue that chunking is a result of optimizing the trade-off between reward and conditional policy complexity. Chunking compresses policies when there is temporal structure in the environment that can be leveraged for action selection, reducing the amount of memory necessary to encode the policy. We experimentally confirm our model’s predictions, showing that chunking reduces conditional policy complexity and reaction times. Chunking also increases with working memory load, consistent with the hypothesis that the degree of policy compression scales with the scarcity of cognitive resources. Finally, chunking also reduces overall working memory load, freeing cognitive resources for the benefit of other, not-chunked information.

2024

1. Learning Dynamics and the Geometry of Neural Dynamics in Recurrent Neural Controllers

   Ann Huang, Satpreet Harcharan Singh, and Kanaka Rajan

   In InterpPol @ RLC 2024 Workshop, 2024
2. Stochastic Wiring of Cell Types Enhances Fitness by Generating Phenotypic Variability

   Divyansha Lachi^*, Ann Huang^*, Augustine N. Mavor-Parker, and 3 more authors

   bioRxiv, 2024

   Abs DOI

   The development of neural connectivity is a crucial biological process that gives rise to diverse brain circuits and behaviors. Neural development is a stochastic process, but this stochasticity is often treated as a nuisance to overcome rather than as a functional advantage. Here we use a computational model, in which connection probabilities between discrete cell types are genetically specified, to investigate the benefits of stochasticity in the development of neural wiring. We show that this model can be viewed as a generalization of a powerful class of artificial neural networks—Bayesian neural networks—where each network parameter is a sample from a distribution. Our results reveal that stochasticity confers a greater benefit in large networks and variable environments, which may explain its role in organisms with larger brains. Surprisingly, we find that the average fitness over a population of agents is higher than a single agent defined by the average connection probability. Our model reveals how developmental stochasticity, by inducing a form of non-heritable phenotypic variability, can increase the probability that at least some individuals will survive in rapidly changing, unpredictable environments. Our results suggest how stochasticity may be an important feature rather than a bug in neural development.Competing Interest StatementThe authors have declared no competing interest.

2023

1. 

   Latent exploration for Reinforcement Learning

   Alberto Silvio Chiappa, Alessandro Marin Vargas, Ann Huang, and 1 more author

   In Advances in Neural Information Processing Systems, 2023

2. 

   Temporal encoding in deep reinforcement learning agents

   Dongyan Lin, Ann Huang, and Blake Aaron Richards

   Scientific Reports, 2023

   DOI