no code implementations • 15 Mar 2024 • Priyanka Sukumaran, Conor Houghton, Nina Kazanina
Language models were tasked with learning the gender of a novel noun embedding from a few examples in one grammatical agreement context and predicting agreement in another, unseen context.
1 code implementation • 6 Sep 2023 • James Malkin, Cian O'Donnell, Conor Houghton, Laurence Aitchison
The resulting networks revealed a tradeoff between circuit performance and the energetic cost of synaptic reliability.
no code implementations • 9 May 2023 • Daniel Collins, Conor Houghton, Nirav Ajmeri
Human social behavior is influenced by individual differences in social preferences.
no code implementations • 28 Feb 2023 • Conor Houghton, Nina Kazanina, Priyanka Sukumaran
Large language models are not detailed models of human linguistic processing.
1 code implementation • 31 Oct 2022 • Priyanka Sukumaran, Conor Houghton, Nina Kazanina
However, we do not have a mechanistic understanding of how LSTMs perform such linguistic tasks.
1 code implementation • 7 Jul 2022 • Davide Turco, Conor Houghton
Bayesian hierarchical models are well-suited to analyzing the often noisy data from electroencephalography experiments in cognitive neuroscience: these models provide an intuitive framework to account for structures and correlations in the data, and they allow a straightforward handling of uncertainty.
no code implementations • ICLR 2019 • Ivan Chelombiev, Conor Houghton, Cian O'Donnell
With two improved methods of estimation, firstly, we show that saturation of the activation function is not required for compression, and the amount of compression varies between different activation functions.