Giancarlo Kerg

发表

Non-normal Recurrent Neural Network (nnRNN): learning long time dependencies while improving expressivity with transient dynamics

Yoshua Bengio, Gauthier Gidel, Giancarlo Kerg, 2019, NeurIPS.

Untangling tradeoffs between recurrence and self-attention in artificial neural networks

Yoshua Bengio, Anirudh Goyal, Giancarlo Kerg, 2020, NeurIPS.

Goal-driven optimization of single-neuron properties in artiﬁcial networks reveals regularization role of neural diversity and adaptation in the brain

Guy Wolf, Guillaume Lajoie, Giancarlo Kerg, 2022 .

I NDUCTIVE B IASES FOR R ELATIONAL T ASKS

Blake A. Richards, D. Rolnick, Y. Bengio, 2022 .

On Neural Architecture Inductive Biases for Relational Tasks

Blake A. Richards, D. Rolnick, Y. Bengio, 2022, ArXiv.

Goal-driven optimization of single-neuron properties in artificial networks reveals regularization role of neural diversity and adaptation in the brain

Guy Wolf, Guillaume Lajoie, Giancarlo Kerg, 2022, bioRxiv.

Continuous-Time Meta-Learning with Forward Mode Differentiation

Pierre-Luc Bacon, Y. Bengio, T. Deleu, 2022, ICLR.

Catastrophic Fisher Explosion: Early Phase Fisher Matrix Impacts Generalization pdf

Richard Socher, Kyunghyun Cho, Krzysztof J. Geras, 2020, ICML.

Advantages of biologically-inspired adaptive neural activation in RNNs during learning pdf

Giancarlo Kerg, Guy Wolf, Guillaume Lajoie, 2020, ArXiv.

Neural networks with optimized single-neuron adaptation uncover biologically plausible regularization

Guy Wolf, Guillaume Lajoie, Giancarlo Kerg, 2023, bioRxiv.

h-detach: Modifying the LSTM Gradient Towards Better Optimization pdf

Ioannis Mitliagkas, Yoshua Bengio, Devansh Arpit, 2018, ICLR.

Untangling tradeoffs between recurrence and self-attention in neural networks pdf

Yoshua Bengio, Giancarlo Kerg, Guillaume Lajoie, 2020, ArXiv.

Learning Long-term Dependencies Using Cognitive Inductive Biases in Self-attention RNNs

Giancarlo Kerg, Yoshua Bengio, Guillaume Lajoie, 2020 .