2 papers
cs.LG2021
A Meta-Learned Neuron model for Continual Learning
Rodrigue Siry
Continual learning is the ability to acquire new knowledge without forgetting the previously learned one, assuming no further access to past training data. Neural network approxima…
cs.LG2021
On the inductive biases of deep domain adaptation
Rodrigue Siry, Louis Hémadou, Loïc Simon +1
Domain alignment is currently the most prevalent solution to unsupervised domain-adaptation tasks and are often being presented as minimizers of some theoretical upper-bounds on ri…