`

Timezone: »

 
Poster
Leader Stochastic Gradient Descent for Distributed Training of Deep Learning Models
Yunfei Teng · Wenbo Gao · François Chalus · Anna Choromanska · Donald Goldfarb · Adrian Weller

Thu Dec 12 10:45 AM -- 12:45 PM (PST) @ East Exhibition Hall B + C #166
in Deep Learning -- Optimization for Deep Networks »

We consider distributed optimization under communication constraints for training deep learning models. We propose a new algorithm, whose parameter updates rely on two forces: a regular gradient step, and a corrective direction dictated by the currently best-performing worker (leader). Our method differs from the parameter-averaging scheme EASGD in a number of ways: (i) our objective formulation does not change the location of stationary points compared to the original optimization problem; (ii) we avoid convergence decelerations caused by pulling local workers descending to different local minima to each other (i.e. to the average of their parameters); (iii) our update by design breaks the curse of symmetry (the phenomenon of being trapped in poorly generalizing sub-optimal solutions in symmetric non-convex landscapes); and (iv) our approach is more communication efficient since it broadcasts only parameters of the leader rather than all workers. We provide theoretical analysis of the batch version of the proposed algorithm, which we call Leader Gradient Descent (LGD), and its stochastic variant (LSGD). Finally, we implement an asynchronous version of our algorithm and extend it to the multi-leader setting, where we form groups of workers, each represented by its own local leader (the best performer in a group), and update each worker with a corrective direction comprised of two attractive forces: one to the local, and one to the global leader (the best performer among all workers). The multi-leader setting is well-aligned with current hardware architecture, where local workers forming a group lie within a single computational node and different groups correspond to different nodes. For training convolutional neural networks, we empirically demonstrate that our approach compares favorably to state-of-the-art baselines.

Keywords: Non-Convex Optimization · Optimization for Deep Networks · Deep Learning · Optimization
[ Paper]  [ Poster] 

Author Information

Yunfei Teng (New York University)
Wenbo Gao (Columbia University)
François Chalus (Credit Suisse)
Anna Choromanska (NYU)
Donald Goldfarb (Columbia University)
Adrian Weller (Cambridge, Alan Turing Institute)

Adrian Weller is Programme Director for AI at The Alan Turing Institute, the UK national institute for data science and AI, where he is also a Turing Fellow leading work on safe and ethical AI. He is a Principal Research Fellow in Machine Learning at the University of Cambridge, and at the Leverhulme Centre for the Future of Intelligence where he is Programme Director for Trust and Society. His interests span AI, its commercial applications and helping to ensure beneficial outcomes for society. He serves on several boards including the Centre for Data Ethics and Innovation. Previously, Adrian held senior roles in finance.

More from the Same Authors