TabQL: In-Context Q-Learning with Tabular Foundation Models
For reinforcement learning practitioners, TabQL offers a new approach to Q-learning that leverages in-context learning to reduce the need for extensive retraining, though it is incremental as it builds on DQN.
TabQL replaces the parametric Q-network in DQN with a tabular foundation model that uses in-context learning for Q-value inference, achieving improved sample efficiency by amortizing Bellman updates. Experiments on benchmarks demonstrate effectiveness.
We propose Tabular Q-Learning (TabQL), a reinforcement learning framework that replaces the conventional parametric Q-network in Deep Q-Learning (DQN) with a tabular foundation model endowed with in-context learning capabilities. The key idea is to represent Q-values through a sequence-to-sequence foundation model operating over a tabularized representation of state-action-Q-value tuples, enabling rapid adaptation from limited online interaction by conditioning on recent experience. TabQL departs from classical DQN by leveraging (i) zero- or few-shot Q-value inference via in-context updates, and (ii) a warm-up phase using standard DQN to bootstrap high-quality context. Particularly, to enhance the context quality, new transitions are generated by executing actions output by TabQL with predicted Q values from DQN. We formalize TabQL, analyze its convergence and sample complexity under mild assumptions, and show that TabQL interpolates between vanilla Q-learning and DQN with in-context learning. Our analysis demonstrates that TabQL achieves improved efficiency compared to DQN by amortizing Bellman updates through in-context learning. Extensive numerical experiments with several benchmarks showcase the effectiveness and efficacy of the proposed TabQL.