Acceleration by Stepsize Hedging I: Multi-Step Descent and the Silver Stepsize Schedule
share
Can we accelerate convergence of gradient descent without changing the algorithm – just by carefully choosing stepsizes? Surprisingly, we show that the answer is yes. Our proposed Silver Stepsize Schedule optimizes strongly convex functions in k^log_ρ 2≈ k^0.7864 iterations, where ρ=1+√(2) is the silver ratio and k is the condition number. This is intermediate between the textbook unaccelerated rate k and the accelerated rate √(k) due to Nesterov in 1983. The non-strongly convex setting is conceptually identical, and standard black-box reductions imply an analogous accelerated rate ε^-log_ρ 2≈ε^-0.7864. We conjecture and provide partial evidence that these rates are optimal among all possible stepsize schedules. The Silver Stepsize Schedule is constructed recursively in a fully explicit way. It is non-monotonic, fractal-like, and approximately periodic of period k^log_ρ 2. This leads to a phase transition in the convergence rate: initially super-exponential (acceleration regime), then exponential (saturation regime).
READ FULL TEXT
