Making Models Shallow Again: Jointly Learning to Reduce Non-Linearity and Depth for Latency-Efficient Private Inference.
Souvik KunduYuke ZhangDake ChenPeter A. BeerelPublished in: CVPR Workshops (2023)
Keyphrases
- structured prediction
- efficient learning
- dynamic bayesian networks
- learning process
- online learning
- reinforcement learning
- accurate models
- prior knowledge
- active learning
- learned models
- learning models
- probabilistic inference
- boltzmann machine
- undirected graphical models
- inference process
- decision theoretic
- inductive inference
- machine learning
- bayesian framework
- learning problems
- unsupervised learning
- response time
- learning algorithm