DGD^2: A Linearly Convergent Distributed Algorithm For High-dimensional Statistical Recovery.
Marie MarosGesualdo ScutariPublished in: NeurIPS (2022)
Keyphrases
- high dimensional
- optimization algorithm
- learning algorithm
- dynamic programming
- improved algorithm
- times faster
- worst case
- k means
- preprocessing
- optimal solution
- objective function
- computational complexity
- neural network
- theoretical analysis
- nearest neighbor
- similarity measure
- simulated annealing
- multi dimensional
- distributed environment
- decision trees
- image segmentation
- expectation maximization
- distributed systems
- high accuracy
- experimental evaluation
- search space
- significant improvement
- tree structure
- multi agent
- convergence rate
- genetic algorithm