Get the latest news, advances in research, policy work, and education program updates from HAI in your inbox weekly.
Sign Up For Latest News
While Large Language Models (LLMs) show promise in many domains, relying on them for direct policy generation in games often results in illegal moves and poor strategic play.
An AI Measurement Science (AIMS) Seminar Series
An AI Measurement Science (AIMS) Seminar Series
Clinical Mind AI is a Stanford-developed, international collaborative platform designed to strengthen clinical reasoning education through AI-simulated patient encounters across health professions.
Clinical Mind AI is a Stanford-developed, international collaborative platform designed to strengthen clinical reasoning education through AI-simulated patient encounters across health professions.
Our November session is Scaling Training Across GPUs and Nodes with speaker Zoe Ryan. This session focuses on efficient strategies for using multiple GPUs and nodes. We will discuss how to deploy strategies of data parallelism and model parallelism to scale to multiple GPUs through PyTorch, enabling faster training times and better model performance.
Location: CoDa W401 & Zoom
Refreshments and snacks provided
Please follow the link to register
Our November session is Scaling Training Across GPUs and Nodes with speaker Zoe Ryan. This session focuses on efficient strategies for using multiple GPUs and nodes. We will discuss how to deploy strategies of data parallelism and model parallelism to scale to multiple GPUs through PyTorch, enabling faster training times and better model performance.
Location: CoDa W401 & Zoom
Refreshments and snacks provided
Please follow the link to register
In this talk, I present an approach that moves away from direct prompting, instead using LLMs as program synthesizers to bridge the gap between natural language rules and symbolic world models. The LLM receives a game description and example trajectories, and outputs an executable, symbolic world model (CWM) represented in Python. The trajectories also ensure the rules are correctly captured and aid in refining the CWM if they are not. Note that even trajectories containing only a single player's observations and actions can be used to help validate and refine CWMs. Furthermore, partially observed trajectories also allow comparisons between CWMs via a bound on the likelihood.
Given a CWM, Monte Carlo Tree Search (MCTS) or Reinforcement Learning (RL) methods can play the game, and gameplay can be further enhanced by adding in LLM-derived synthesized value functions. Imperfect information games are handled by having the LLM synthesize inference functions to impute information sets, or by directly training reinforcement learning policies on top of the CWM.
