MARS: Masked Automatic Ranks Selection in Tensor Decompositions
arXiv:2006.10859
Abstract
Tensor decomposition methods have proven effective in various applications, including compression and acceleration of neural networks. At the same time, the problem of determining optimal decomposition ranks, which present the crucial parameter controlling the compression-accuracy trade-off, is still acute. In this paper, we introduce MARS -- a new efficient method for the automatic selection of ranks in general tensor decompositions. During training, the procedure learns binary masks over decomposition cores that "select" the optimal tensor structure. The learning is performed via relaxed maximum a posteriori (MAP) estimation in a specific Bayesian model and can be naturally embedded into the standard neural network training routine. Diverse experiments demonstrate that MARS achieves better results compared to previous works in various tasks.
arXiv commentAISTATS 2023
References in corpus (12)
- PyTorch: An Imperative Style, High-Performance Deep Learning Library
- The Concrete Distribution: A Continuous Relaxation of Discrete Random Variables
- Tensorizing Neural Networks
- Tensor Ring Decomposition
- On the Power of Over-parametrization in Neural Networks with Quadratic Activation
- Ultimate tensorization: compressing convolutional and FC layers alike
- A Tensorized Transformer for Language Modeling
- Tensorized Embedding Layers for Efficient Model Compression
- Wide Compression: Tensor Ring Nets
- Efficient tensor completion: Low-rank tensor train
- Bayesian Tensorized Neural Networks with Automatic Rank Selection
- Adaptive Learning of Tensor Network Structures