Abstract | ||
---|---|---|
The diffusion strategy for distributed learning from streaming data employs local stochastic gradient updates along with exchange of iterates over neighborhoods. In this work we establish that agents cluster around a network centroid in the mean-fourth sense and proceeded to study the dynamics of this point. We establish expected descent in non-convex environments in the large-gradient regime and introduce a short-term model to examine the dynamics over finite-time horizons. Using this model, we establish that the diffusion strategy is able to escape from strict saddle-points in O(1/μ) iterations, where μ denotes the step-size; it is also able to return approximately second-order stationary points in a polynomial number of iterations. Relative to prior works on the polynomial escape from saddle-points, most of which focus on centralized perturbed or stochastic gradient descent, our approach requires less restrictive conditions on the gradient noise process. |
Year | DOI | Venue |
---|---|---|
2019 | 10.1109/CAMSAP45676.2019.9022458 | 2019 IEEE 8th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP) |
Keywords | DocType | ISBN |
Stochastic optimization,adaptation,nonconvex costs,saddle point,escape time,gradient noise,stationary points,distributed optimization,diffusion learning | Conference | 978-1-7281-5550-0 |
Citations | PageRank | References |
0 | 0.34 | 15 |
Authors | ||
2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Stefan Vlaski | 1 | 23 | 11.39 |
Ali H. Sayed | 2 | 9134 | 667.71 |