On architectural choices in deep learning: From network structure to gradient convergence and parameter estimation.
Vamsi K. IthapuSathya N. RaviVikas SinghPublished in: CoRR (2017)
Keyphrases
- parameter estimation
- network structure
- deep learning
- model selection
- maximum likelihood
- markov random field
- unsupervised learning
- structure learning
- least squares
- social networks
- machine learning
- em algorithm
- parameter estimation algorithm
- expectation maximization
- approximate inference
- mental models
- graph cuts
- image processing
- weakly supervised
- reinforcement learning