Flattened one-bit stochastic gradient descent: compressed distributed optimization with controlled variance.
Alexander StollenwerkLaurent JacquesPublished in: CoRR (2024)
Keyphrases
- stochastic gradient descent
- stochastic gradient
- least squares
- loss function
- matrix factorization
- step size
- random forests
- weight vector
- optimization algorithm
- support vector machine
- importance sampling
- regularization parameter
- collaborative filtering
- markov chain
- image compression
- convergence rate
- multiresolution
- multiple kernel learning
- online algorithms