In this work, we study the convergence in high probability
of clipped gradient methods when the noise distribution has heavy
tails, i.e., with bounded $p$th moments, for some $1< p \leq 2$. Prior
works in this setting follow the same recipe of using concentration
inequalities and an inductive argument with union bound to bound the
iterates across all iterations. This method results in an increase
in the failure probability by a factor of $T$, where $T$ is the
number of iterations. We instead propose a new analysis approach based
on bounding the moment generating function of a well chosen supermartingale
sequence. We improve the dependency on $T$ in the convergence guarantee
for a wide range of algorithms with clipped gradients, including stochastic
(accelerated) mirror descent for convex objectives and stochastic
gradient descent for nonconvex objectives. Our high probability bounds
achieve the optimal convergence rates and match the best currently
known in-expectation bounds. Our approach naturally allows the algorithms
to use time-varying step sizes and clipping parameters when the time
horizon is unknown, which appears difficult or even impossible using
existing techniques from prior works. Furthermore, we show that in
the case of clipped stochastic mirror descent, several problem constants,
including the initial distance to the optimum, are not required when
setting step sizes and clipping parameters.
more »
« less
Momentum based stochastic gradient methods such as heavy ball (HB) and Nesterov's accelerated gradient descent (NAG) method are widely used in practice for training deep networks and other supervised learning models, as they often provide significant improvements over stochastic gradient descent (SGD). Rigorously speaking, fast gradient methods have provable improvements over gradient descent only for the deterministic case, where the gradients are exact. In the stochastic case, the popular explanations for their wide applicability is that when these fast gradient methods are applied in the stochastic case, they partially mimic their exact gradient counterparts, resulting in some practical gain. This work provides a counterpoint to this belief by proving that there exist simple problem instances where these methods cannot outperform SGD despite the best setting of its parameters. These negative problem instances are, in an informal sense, generic; they do not look like carefully constructed pathological instances. These results suggest (along with empirical evidence) that HB or NAG's practical performance gains are a by-product of minibatching.
Furthermore, this work provides a viable (and provable) alternative, which, on the same set of problem instances, significantly improves over HB, NAG, and SGD's performance. This algorithm, referred to as Accelerated Stochastic Gradient Descent (ASGD), is a simple to implement stochastic algorithm, based on a relatively less popular variant of Nesterov's Acceleration. Extensive empirical results in this paper show that ASGD has performance gains over HB, NAG, and SGD. The code for implementing the ASGD Algorithm can be found at https://github.com/rahulkidambi/AccSGD.
more »
« less
We study performance of accelerated first-order optimization algorithms in the presence of additive white stochastic disturbances. For strongly convex quadratic problems, we explicitly evaluate the steady-state variance of the optimization variable in terms of the eigenvalues of the Hessian of the objective function. We demonstrate that, as the condition number increases, variance amplification of both Nesterov's accelerated method and the heavy-ball method by Polyak is significantly larger than that of the standard gradient descent. In the context of distributed computation over networks, we examine the role of network topology and spatial dimension on the performance of these first-order algorithms. For d-dimensional tori, we establish explicit asymptotic dependence for the variance amplification on the network size and the corresponding condition number. Our results demonstrate detrimental influence of acceleration on amplification of stochastic disturbances and suggest that metrics other than convergence rate have to be considered when evaluating performance of optimization algorithms.
more »
« less
The theory of integral quadratic constraints (IQCs) allows the certification of exponential convergence of interconnected systems containing nonlinear or uncertain elements. In this work, we adapt the IQC theory to study first-order methods for smooth and strongly-monotone games and show how to design tailored quadratic constraints to get tight upper bounds of convergence rates. Using this framework, we recover the existing bound for the gradient method~(GD), derive sharper bounds for the proximal point method~(PPM) and optimistic gradient method~(OG), and provide for the first time a global convergence rate for the negative momentum method~(NM) with an iteration complexity O(κ1.5), which matches its known lower bound. In addition, for time-varying systems, we prove that the gradient method with optimal step size achieves the fastest provable worst-case convergence rate with quadratic Lyapunov functions. Finally, we further extend our analysis to stochastic games and study the impact of multiplicative noise on different algorithms. We show that it is impossible for an algorithm with one step of memory to achieve acceleration if it only queries the gradient once per batch (in contrast with the stochastic strongly-convex optimization setting, where such acceleration has been demonstrated). However, we exhibit an algorithm which achieves acceleration with two gradient queries per batch.
more »
« less