Lambda

Burkov's new book takes you from the Policy Gradient Theorem to working PyTorch code. You implement REINFORCE from scratch, then move to PPO and Deep Q-Networks, all on the same rocket-landing task so you can compare how each algorithm learns and behaves. It also covers RLHF and verifiable rewards for language models. Lambda is sponsoring students with cloud credits to run the code.

GPU-accelerated simulation is the bridge: embodied AI benchmarks in sim, then sim-to-real transfer for real robot policy learning. Three challenges, BEHAVIOR, RoboEval, RoboChess, make that pipeline concrete. Nov 12, 8:30 AM CST, Room 212, JW Marriott, Austin.

Lambda is sponsoring two workshops at CoRL 2026 in Austin and co-organizing them with NVIDIA. We’re also presenting I3L, a workshop paper with Li Fei-Fei’s lab at Stanford. I3L studies whole-body mobile manipulation through iterative, interactive imitation learning. On JoyLo+, the robot can hand control to a person through touch. The authors tested the approach on three robot embodiments and eight tasks. They report a 68.3 point gain in average autonomous success over the base policies, at least 90% success on every task, and a 72% reduction in average takeover time. Both workshops are Nov…