We study the decentralized distributed computation of discrete approximations for the regularized Wasserstein barycenter of a finite set of continuous probability measures distributedly stored over a network. We assume there is a network of agents/machines/computers, and each agent holds a private continuous probability measure and seeks to compute the barycenter of all the measures in the network by getting samples from its local measure and exchanging information with its neighbors. Motivated by this problem, we develop, and analyze, a novel accelerated primal-dual stochastic gradient method for general stochastic convex optimization problems with linear equality constraints. Then, we apply this method to the decentralized distributed optimization setting to obtain a new algorithm for the distributed semi-discrete regularized Wasserstein barycenter problem. Moreover, we show explicit non-asymptotic complexity for the proposed algorithm. Finally, we show the effectiveness of our method on the distributed computation of the regularized Wasserstein barycenter of univariate Gaussian and von Mises distributions, as well as some applications to image aggregation.

We study private stochastic convex optimization (SCO) under user-level differential privacy (DP) constraints. In this setting, there are n users (e.g., cell phones), each possessing m data items (e.g., text messages), and we need to protect the privacy of each user's entire collection of data items. Existing algorithms for user-level DP SCO are impractical in many large-scale machine learning scenarios because: (i) they make restrictive assumptions on the smoothness parameter of the loss function and require the number of users to grow polynomially with the dimension of the parameter space; or (ii) they are prohibitively slow, requiring at least (mn) {3/2} gradient computations for smooth losses and (mn) 3 computations for non-smooth losses. To address these limitations, we provide novel user-level DP algorithms with state-of-the-art excess risk and runtime guarantees, without stringent assumptions. First, we develop a linear-time algorithm with state-of-the-art excess risk (for a non-trivial linear-time algorithm) under a mild smoothness assumption.

From my understanding, the paper makes a good technical contribution, unifying a large body of work on isotonic regression (IR). The basic idea seems intuitive, and is to employ techniques from the fast solvers of linear systems. Thus, from the perspective of novelty and technical content, I cannot raise any issues (based on my limited understanding -- regrettably, I do not have the background to check the proofs). But my concern with the paper is simply that it may be better suited to a algorithms/theoretical CS conference or journal, such as those where the work it improves upon ([16] -- [20]), and the work it employs in developing the algorithm ([21] -- [29]) were published. It is unclear to me whether the results in the paper would be of sufficient interest to the broader NIPS community. In particular: - while IR has seen some interesting applications to learning problems of late, it is not (in my estimation) a core ML tool for which a faster algorithm is by itself of wide interest.

We propose two novel algorithms for constructing convex collision-free polytopes in robot configuration space. Finding these polytopes enables the application of stronger motion-planning frameworks such as trajectory optimization with Graphs of Convex Sets [1] and is currently a major roadblock in the adoption of these approaches. In this paper, we build upon IRIS-NP (Iterative Regional Inflation by Semidefinite & Nonlinear Programming) [2] to significantly improve tunability, runtimes, and scaling to complex environments. IRIS-NP uses nonlinear programming paired with uniform random initialization to find configurations on the boundary of the free configuration space. Our key insight is that finding near-by configuration-space obstacles using sampling is inexpensive and greatly accelerates region generation. We propose two algorithms using such samples to either employ nonlinear programming more efficiently (IRIS-NP2) or circumvent it altogether using a massively-parallel zero-order optimization strategy (IRIS-ZO). We also propose a termination condition that controls the probability of exceeding a user-specified permissible fraction-in-collision, eliminating a significant source of tuning difficulty in IRIS-NP. We compare performance across eight robot environments, showing that IRIS-ZO achieves an order-of-magnitude speed advantage over IRIS-NP. IRIS-NP2, also significantly faster than IRIS-NP, builds larger polytopes using fewer hyperplanes, enabling faster downstream computation.

The authors first show a negative result that soft-policy updates using the multi-step greedy policies do not guarantee policy improvement. Then the authors proposed an algorithm that uses cautious soft updates (only update to the kappa greedy policy only when assured to improve, otherwise stay with one-step greedy policy) and show that it converges to the optimal policy. Lastly the authors studied hard updates by extending APIs to multi-step greedy policy setting. Comments: 1. Theorem 2 presents an interesting and surprising result. Though the authors presented the example in the proof sketch, but I wonder if the authors could provide more intuitions behind this? Based on the theorem, for multi-step greedy policy, it seems that h needs to be bigger than 2. So I suspect that h 2 will still work (meaning there could exist small alpha)? Obviously h 1 works, but then why when h 3, the soft-update suddenly stops working unless alpha is exactly equal to 1? I would expect that one would require larger alpha when h gets larger.

This paper presents a distributed algorithm for computing Wasserstein barycenters. The basic setup is that each agent in the decentralized system has access to one probability distribution; similar to "gossip" based optimization techniques in the classical case (e.g. It seems this paper missed the closest related work, "Stochastic Wasserstein Barycenters" (Claici et al., ArXiv/ICML), which proposes a nonconvex semidiscrete barycenter optimization algorithm. Certainly any final version of this paper needs to compare to that work carefully. It may also be worth noting that the Wasserstein propagation algorithm in "Convolutional Wasserstein Distances: Efficient Optimal Transportation on Geometric Domains" (2015) could be implemented easily on a network in a similar fashion to what is proposed in this paper; see their Algorithm 4. Like lots of previous work in OT, this technique uses entropic regularization to make transport tractable; they solve the smoothed dual.

We study the decentralized distributed computation of discrete approximations for the regularized Wasserstein barycenter of a finite set of continuous probability measures distributedly stored over a network. We assume there is a network of agents/machines/computers, and each agent holds a private continuous probability measure and seeks to compute the barycenter of all the measures in the network by getting samples from its local measure and exchanging information with its neighbors. Motivated by this problem, we develop, and analyze, a novel accelerated primal-dual stochastic gradient method for general stochastic convex optimization problems with linear equality constraints. Then, we apply this method to the decen- tralized distributed optimization setting to obtain a new algorithm for the distributed semi-discrete regularized Wasserstein barycenter problem. Moreover, we show explicit non-asymptotic complexity for the proposed algorithm.

Now a days, the major challenge in machine learning is the `Big~Data' challenge. The big data problems due to large number of data points or large number of features in each data point, or both, the training of models have become very slow. The training time has two major components: Time to access the data and time to process (learn from) the data. In this paper, we have proposed one possible solution to handle the big data problems in machine learning. The idea is to reduce the training time through reducing data access time by proposing systematic sampling and cyclic/sequential sampling to select mini-batches from the dataset. To prove the effectiveness of proposed sampling techniques, we have used Empirical Risk Minimization, which is commonly used machine learning problem, for strongly convex and smooth case. The problem has been solved using SAG, SAGA, SVRG, SAAG-II and MBSGD (Mini-batched SGD), each using two step determination techniques, namely, constant step size and backtracking line search method. Theoretical results prove the same convergence for systematic sampling, cyclic sampling and the widely used random sampling technique, in expectation. Experimental results with bench marked datasets prove the efficacy of the proposed sampling techniques.