Portrait de Shenyang Huang

Shenyang Huang

Collaborateur·rice alumni - McGill
Superviseur⋅e principal⋅e
Reihaneh Rabbany
Co-supervisor
Guillaume Rabusseau
Sujets de recherche
Apprentissage profond
Apprentissage spectral
Exploration des données
Modélisation moléculaire
Réseaux de neurones en graphes
Site web
Google Scholar
GitHub

Publications

Neural Architecture Search for Class-incremental Learning
Shenyang Huang
Vincent François-Lavet
Guillaume Rabusseau
In class-incremental learning, a model learns continuously from a sequential data stream in which new classes occur. Existing methods often … (voir plus)rely on static architectures that are manually crafted. These methods can be prone to capacity saturation because a neural network's ability to generalize to new concepts is limited by its fixed capacity. To understand how to expand a continual learner, we focus on the neural architecture design problem in the context of class-incremental learning: at each time step, the learner must optimize its performance on all classes observed so far by selecting the most competitive neural architecture. To tackle this problem, we propose Continual Neural Architecture Search (CNAS): an autoML approach that takes advantage of the sequential nature of class-incremental learning to efficiently and adaptively identify strong architectures in a continual learning setting. We employ a task network to perform the classification task and a reinforcement learning agent as the meta-controller for architecture search. In addition, we apply network transformations to transfer weights from previous learning step and to reduce the size of the architecture search space, thus saving a large amount of computational resources. We evaluate CNAS on the CIFAR-100 dataset under varied incremental learning scenarios with limited computational power (1 GPU). Experimental results demonstrate that CNAS outperforms architectures that are optimized for the entire dataset. In addition, CNAS is at least an order of magnitude more efficient than naively using existing autoML methods.
2019-09-14
ArXiv (prépublication)
arxiv.org
arxiv.org