Roman Pogodin
Roman Pogodin
PostDoc, McGill/Mila
Verified email at - Homepage
Cited by
Cited by
Self-supervised learning with kernel dependence maximization
Y Li*, R Pogodin*, DJ Sutherland, A Gretton
Advances in Neural Information Processing Systems 34, 2021
Kernelized information bottleneck leads to biologically plausible 3-factor Hebbian learning in deep networks
R Pogodin, P Latham
Advances in Neural Information Processing Systems 33, 2020
Towards biologically plausible convolutional networks
R Pogodin, Y Mehta, T Lillicrap, P Latham
Advances in Neural Information Processing Systems 34, 2021
On First-Order Bounds, Variance and Gap-Dependent Bounds for Adversarial Bandits
R Pogodin, T Lattimore
The Conference on Uncertainty in Artificial Intelligence (UAI) 2019, 2019
Efficient Rank Minimization to Tighten Semidefinite Programming for Unconstrained Binary Quadratic Optimization
R Pogodin, M Krechetov, Y Maximov
2017 55th Annual Allerton Conference on Communication, Control, and …, 2017
Efficient conditionally invariant representation learning
R Pogodin*, N Deka*, Y Li*, DJ Sutherland, V Veitch, A Gretton
ICLR 2023 notable top 5%, 2022
Synaptic Weight Distributions Depend on the Geometry of Plasticity
R Pogodin*, J Cornford*, A Ghosh, G Gidel, G Lajoie, B Richards
arXiv preprint arXiv:2305.19394, 2023
Locally connected networks as ventral stream models
R Pogodin, PE Latham
Brain-Score Workshop, 2022
Working memory facilitates reward-modulated Hebbian learning in recurrent neural networks
R Pogodin, D Corneil, A Seeholzer, J Heng, W Gerstner
NeurIPS 2019 workshop "Real Neurons & Hidden Units: Future directions at the …, 2019
Deep Learning Models of Learning in the Brain
R Pogodin
UCL (University College London), 2023
Quadratic Programming Approach to Fit Protein Complexes into Electron Density Maps
R Pogodin, A Katrutsa, S Grudinin
Information Technology and Systems 2016, 2017
The system can't perform the operation now. Try again later.
Articles 1–11