AI/ reinforcement learning · machine learning · ai research · foundation models

A Regression Model That Plays Atari - No RL Training Required

Researchers show a foundation model trained purely on regression tasks can solve reinforcement learning benchmarks without any additional fine-tuning.

A pre-trained regression model can tackle reinforcement learning problems without ever being trained on RL tasks.

Researchers introduced ICR-RL, a gradient-free method that repurposes a foundation model built for regression - specifically TabPFN, trained on a wide range of tabular regression tasks - and applies it directly to RL control problems. The trick is treating RL as a regression problem, a classical theoretical reduction that has rarely been exploited at the foundation model scale. Tested against Gymnasium's classic-control benchmark, ICR-RL held its own against established RL algorithms including DQN, PPO, and TRPO. No fine-tuning, no RL-specific environment data, no additional training loop.

The result matters because building diverse RL training environments is expensive and brittle - it's one of the main reasons scaling RL foundation models has lagged behind language and vision. If a general-purpose regression model can sidestep that bottleneck, it changes the cost calculus for anyone trying to apply RL to new domains. The finding also raises a quieter question: how much of what we attribute to RL-specific training is actually just competent function approximation?

Classic-control benchmarks like CartPole and Pendulum are the "hello world" of RL - the real test will be whether this approach survives contact with sparse rewards, partial observability, and environments that punish the kind of smooth interpolation regression models love.

TR

The Revision

Written by an AI system from the public sources credited above. How we write →