Zobrazeno 1 - 10
of 19
pro vyhledávání: '"Pogodin, Roman"'
We describe a data-efficient, kernel-based approach to statistical testing of conditional independence. A major challenge of conditional independence testing, absent in tests of unconditional independence, is to obtain the correct test level (the spe
Externí odkaz:
http://arxiv.org/abs/2402.13196
Autor:
Pogodin, Roman, Cornford, Jonathan, Ghosh, Arna, Gidel, Gauthier, Lajoie, Guillaume, Richards, Blake
Publikováno v:
The Twelfth International Conference on Learning Representations, 2024
A growing literature in computational neuroscience leverages gradient descent and learning algorithms that approximate it to study synaptic plasticity in the brain. However, the vast majority of this work ignores a critical underlying assumption: the
Externí odkaz:
http://arxiv.org/abs/2305.19394
Autor:
Pogodin, Roman, Deka, Namrata, Li, Yazhe, Sutherland, Danica J., Veitch, Victor, Gretton, Arthur
Publikováno v:
The Eleventh International Conference on Learning Representations, 2023
We introduce the Conditional Independence Regression CovariancE (CIRCE), a measure of conditional independence for multivariate continuous-valued variables. CIRCE applies as a regularizer in settings where we wish to learn neural features $\varphi(X)
Externí odkaz:
http://arxiv.org/abs/2212.08645
Convolutional networks are ubiquitous in deep learning. They are particularly useful for images, as they reduce the number of parameters, reduce training time, and increase accuracy. However, as a model of the brain they are seriously problematic, si
Externí odkaz:
http://arxiv.org/abs/2106.13031
We approach self-supervised learning of image representations from a statistical dependence perspective, proposing Self-Supervised Learning with the Hilbert-Schmidt Independence Criterion (SSL-HSIC). SSL-HSIC maximizes dependence between representati
Externí odkaz:
http://arxiv.org/abs/2106.08320
Autor:
Pogodin, Roman, Latham, Peter E.
The state-of-the art machine learning approach to training deep neural networks, backpropagation, is implausible for real neural networks: neurons need to know their outgoing weights; training alternates between a bottom-up forward pass (computation)
Externí odkaz:
http://arxiv.org/abs/2006.07123
Reservoir computing is a powerful tool to explain how the brain learns temporal sequences, such as movements, but existing learning schemes are either biologically implausible or too inefficient to explain animal performance. We show that a network c
Externí odkaz:
http://arxiv.org/abs/1910.10559
Autor:
Pogodin, Roman, Lattimore, Tor
We make three contributions to the theory of k-armed adversarial bandits. First, we prove a first-order bound for a modified variant of the INF strategy by Audibert and Bubeck [2009], without sacrificing worst case optimality or modifying the loss es
Externí odkaz:
http://arxiv.org/abs/1903.07890
We propose a method for low-rank semidefinite programming in application to the semidefinite relaxation of unconstrained binary quadratic problems. The method improves an existing solution of the semidefinite programming relaxation to achieve a lower
Externí odkaz:
http://arxiv.org/abs/1708.01690
The paper investigates the problem of fitting protein complexes into electron density maps. They are represented by high-resolution cryoEM density maps converted into overlapping matrices and partly show a structure of a complex. The general purpose
Externí odkaz:
http://arxiv.org/abs/1701.02192