Thomas Mesnard
Thomas Mesnard
DeepMind
Verified email at google.com - Homepage
Title
Cited by
Cited by
Year
Towards biologically plausible deep learning
Y Bengio, DH Lee, J Bornschein, T Mesnard, Z Lin
arXiv preprint arXiv:1502.04156, 2015
2762015
STDP-compatible approximation of backpropagation in an energy-based model
Y Bengio, T Mesnard, A Fischer, S Zhang, Y Wu
Neural computation 29 (3), 555-577, 2017
722017
An objective function for STDP
Y Bengio, T Mesnard, A Fischer, S Zhang, Y Wu
arXiv preprint arXiv:1509.05936, 2015
43*2015
Generalization of equilibrium propagation to vector field dynamics
B Scellier, A Goyal, J Binas, T Mesnard, Y Bengio
arXiv preprint arXiv:1808.04873, 2018
25*2018
Hindsight credit assignment
A Harutyunyan, W Dabney, T Mesnard, M Azar, B Piot, N Heess, ...
arXiv preprint arXiv:1912.02503, 2019
192019
Towards deep learning with spiking neurons in energy based models with contrastive hebbian plasticity
T Mesnard, W Gerstner, J Brea
arXiv preprint arXiv:1612.03214, 2016
132016
From STDP towards Biologically Plausible Deep Learning
Y Bengio, A Fischer, T Mesnard, S Zhang, Y Wu
ICML 2015, Deep Learning Workshop, 2015
42015
Counterfactual Credit Assignment in Model-Free Reinforcement Learning
T Mesnard, T Weber, F Viola, S Thakoor, A Saade, A Harutyunyan, ...
arXiv preprint arXiv:2011.09464, 2020
32020
Geometric Entropic Exploration
ZD Guo, MG Azar, A Saade, S Thakoor, B Piot, BA Pires, M Valko, ...
arXiv preprint arXiv:2101.02055, 2021
22021
Ghost units yield biologically plausible backprop in deep neural networks
T Mesnard, G Vignoud, J Sacramento, W Senn, Y Bengio
arXiv preprint arXiv:1911.08585, 2019
22019
Activation alignment: exploring the use of approximate activity gradients in multilayer networks
T Mesnard, B Richards
Cognitive Computational Neuroscience, 2018
2018
Fully discretized training of neural networks through direct feedback
T Mesnard, G Vignoud, J Binas, Y Bengio
2018
The system can't perform the operation now. Try again later.
Articles 1–12