Provable Benefits of Overparameterization in Model Compression: From Double Descent to Pruning Neural Networks.
Xiangyu ChangYingcong LiSamet OymakChristos ThrampoulidisPublished in: AAAI (2021)
Keyphrases
- neural network
- probabilistic model
- machine learning
- pattern recognition
- bayesian networks
- multiscale
- similarity measure
- artificial neural networks
- hidden markov models
- decision making
- image compression
- computational model
- network structure
- conceptual model
- neural network model
- formal model
- bp neural network
- compression scheme
- hybrid model