Portrait de Daniel Lawson n'est pas disponible

Daniel Lawson

daniel.lawson@mila.quebec
Doctorat - UdeM
Superviseur⋅e principal⋅e
Glen Berseth
Co-supervisor
Khimya Khetarpal
Sujets de recherche
Apprentissage par renforcement
Apprentissage profond
Robotique
Site web
Google Scholar
GitHub

Publications

Imagining Goals with Geometric Horizon Models for Policy Generalization
Daniel Lawson
Dhruv Sreenivas
Adriana Hugessen
Khimya Khetarpal
Glen Berseth
Training from offline data has allowed for substantial progress in domains such as robotics, leading to general-purpose policies that can be… (voir plus) easily applied zero-shot or efficiently finetuned for downstream tasks. However, these policies can still have poor generalization, both due to the choice of modeling objective and from learning from a static dataset. In this work, we focus on the challenging task of zero-shot goal generalization, where a policy is evaluated on unseen tasks that require combining its existing knowledge (compositional generalization). An avenue for improving a policy's generalization is by generating new experience with world models; however, such generation has proven difficult for longer horizons. Thus, to alleviate this issue, we propose TD-Aug, which samples from a geometric horizon model and allows for directly imagining novel outcomes that can be achieved by composing existing knowledge. We demonstrate that training on these future outcomes as goals for goal-conditioned BC policies significantly improves generalization in stitching-based OGBench tasks.
2026-10-12
arXiv (prépublication)
doi.org
arxiv.org
Imagining Goals for Policy Generalization with Geometric Horizon Models
Daniel Lawson
Dhruv Sreenivas
Adriana Hugessen
Khimya Khetarpal
Glen Berseth
Training from offline data has allowed for substantial progress in domains such as robotics, leading to general-purpose policies that can be… (voir plus) easily applied zero-shot or efficiently finetuned for downstream tasks. However, training policies on offline data can lead to poor generalization, both due to the choice of modeling objective and from learning from a static dataset. In this work, we focus on the challenging task of zero-shot goal generalization, where a policy is evaluated on unseen tasks that require reusing its existing knowledge (compositional generalization). An avenue for improving a policy's generalization is through generating new experience through world models; however, such generation has proven difficult for longer horizons. Thus, to alleviate this issue, we propose TD-Aug, sampling from a geometric horizon model, which allows for directly imagining novel outcomes that can be achieved through composing existing knowledge. We demonstrate that training on these future outcomes as goals for goal-conditioned BC and offline RL policies improves generalization in stitching-based OGBench tasks.
2026-07-20
RLBRew @ Reinforcement Learning Conference (publié)
openreview.net
openreview.net
Self-Predictive Representations for Combinatorial Generalization in Behavioral Cloning
Daniel Lawson
Adriana Hugessen
Charlotte Cloutier
Glen Berseth
Khimya Khetarpal
Behavioral cloning (BC) methods trained with supervised learning (SL) are an effective way to learn policies from human demonstrations in do… (voir plus)mains like robotics. Goal-conditioning these policies enables a single generalist policy to capture diverse behaviors contained within an offline dataset. While goal-conditioned behavior cloning (GCBC) methods can perform well on in-distribution training tasks, they do not necessarily generalize zero-shot to tasks that require conditioning on novel state-goal pairs, i.e. combinatorial generalization. In part, this limitation can be attributed to a lack of temporal consistency in the state representation learned by BC; if temporally related states are encoded to similar latent representations, then the out-of-distribution gap for novel state-goal pairs would be reduced. Hence, encouraging this temporal consistency in the representation space should facilitate combinatorial generalization. Successor representations, which encode the distribution of future states visited from the current state, nicely encapsulate this property. However, previous methods for learning successor representations have relied on contrastive samples, temporal-difference (TD) learning, or both. In this work, we propose a simple yet effective representation learning objective,
2025-06-30
rl-conference.cc/RLC/2025/Workshop/RLBrew (publié)
doi.org
openreview.net