167 citations · 191 across the 3 of their papers we have counts for
5 papers
Tracr: Compiled Transformers as a Laboratory for Interpretability
David Lindner, János Kramár, Sebastian Farquhar +3
We show how to "compile" human-readable programs into standard decoder-only transformer models. Our compiler, Tracr, generates models with known structure. This structure can be us…
Sample-based Approximation of Nash in Large Many-Player Games via Gradient Descent
Ian Gemp, Rahul Savani, Marc Lanctot +6
Nash equilibrium is a central concept in game theory. Several Nash solvers exist, yet none scale to normal-form games with many actions and many players, especially those with payo…
Learning to Play No-Press Diplomacy with Best Response Policy Iteration
Thomas Anthony, Tom Eccles, Andrea Tacchetti +11
Recent advances in deep reinforcement learning (RL) have led to considerable progress in many 2-player zero-sum games, such as Go, Poker and Starcraft. The purely adversarial natur…
Should I tear down this wall? Optimizing social metrics by evaluating novel actions
János Kramár, Neil Rabinowitz, Tom Eccles +1
One of the fundamental challenges of governance is deciding when and how to intervene in multi-agent systems in order to impact group-wide metrics of success. This is particularly…
Zoneout: Regularizing RNNs by Randomly Preserving Hidden Activations
David Krueger, Tegan Maharaj, János Kramár +7
We propose zoneout, a novel method for regularizing RNNs. At each timestep, zoneout stochastically forces some hidden units to maintain their previous values. Like dropout, zoneout…