Stochastic Gradient Descent as Approximate Bayesian Inference.
Stephan MandtMatthew D. HoffmanDavid M. BleiPublished in: J. Mach. Learn. Res. (2017)
Keyphrases
- bayesian inference
- stochastic gradient descent
- expectation propagation
- hyperparameters
- least squares
- step size
- probabilistic model
- loss function
- regularization parameter
- prior information
- matrix factorization
- random forests
- importance sampling
- online algorithms
- markov chain monte carlo
- weight vector
- particle filter
- multiple kernel learning
- cost function
- support vector machine
- convergence rate
- monte carlo