2 code implementations • NeurIPS 2023 • Arthur Conmy, Augustine N. Mavor-Parker, Aengus Lynch, Stefan Heimersheim, Adrià Garriga-Alonso
For example, the ACDC algorithm rediscovered 5/5 of the component types in a circuit in GPT-2 Small that computes the Greater-Than operation.
no code implementations • 14 Sep 2022 • Augustine N. Mavor-Parker, Matthew J. Sargent, Andrea Banino, Lewis D. Griffin, Caswell Barry
Consequently, impressive improvements in sample efficiency have been achieved when a suitable MDP homomorphism can be constructed a priori -- usually by exploiting a practioner's knowledge of environment symmetries.
1 code implementation • 8 Feb 2021 • Augustine N. Mavor-Parker, Kimberly A. Young, Caswell Barry, Lewis D. Griffin
Exploration in environments with sparse rewards is difficult for artificial agents.
no code implementations • ICLR 2022 • Jason D. McEwen, Christopher G. R. Wallis, Augustine N. Mavor-Parker
Convolutional neural networks (CNNs) constructed natively on the sphere have been developed recently and shown to be highly effective for the analysis of spherical data.
1 code implementation • ICLR 2021 • Oliver J. Cobb, Christopher G. R. Wallis, Augustine N. Mavor-Parker, Augustin Marignier, Matthew A. Price, Mayeul d'Avezac, Jason D. McEwen
We develop two new strictly equivariant layers with reduced complexity $\mathcal{O}(CL^4)$ and $\mathcal{O}(CL^3 \log L)$, making larger, more expressive models computationally feasible.