Log-Scale Quantization in Distributed First-Order Methods: Gradient-based Learning from Distributed Data

Decentralized strategies are of interest for learning from large-scale data over networks. This paper studies learning over a network of geographically distributed nodes/agents subject to quantization. Each node possesses a private local cost function, collectively contributing to a global cost function, which the proposed methodology aims to minimize. In contrast to many existing literature, the information exchange among nodes is quantized. We adopt a first-order computationally-efficient distributed optimization algorithm (with no extra inner consensus loop) that leverages node-level gradient correction based on local data and network-level gradient aggregation only over nearby nodes. This method only requires balanced networks with no need for stochastic weight design. It can handle log-scale quantized data exchange over possibly time-varying and switching network setups. We analyze convergence over both structured networks (for example, training over data-centers) and ad-hoc multi-agent networks (for example, training over dynamic robotic networks). Through analysis and experimental validation, we show that (i) structured networks generally result in a smaller optimality gap, and (ii) logarithmic quantization leads to smaller optimality gap compared to uniform quantization.

Consensus-based Networked Tracking in Presence of Heterogeneous Time-Delays

We propose a distributed (single) target tracking scheme based on networked estimation and consensus algorithms over static sensor networks. The tracking part is based on linear time-difference-of-arrival (TDOA) measurement proposed in our previous works. This paper, in particular, develops delay-tolerant distributed filtering solutions over sparse data-transmission networks. We assume general arbitrary heterogeneous delays at different links. This may occur in many realistic large-scale applications where the data-sharing between different nodes is subject to latency due to communication-resource constraints or large spatially distributed sensor networks. The solution we propose in this work shows improved performance (verified by both theory and simulations) in such scenarios. Another privilege of such distributed schemes is the possibility to add localized fault-detection and isolation (FDI) strategies along with survivable graph-theoretic design, which opens many follow-up venues to this research. To our best knowledge no such delay-tolerant distributed linear algorithm is given in the existing distributed tracking literature.

Distributed saddle point problems for strongly concave-convex functions

In this paper, we propose GT-GDA, a distributed optimization method to solve saddle point problems of the form: $\min_{\mathbf{x}} \max_{\mathbf{y}} \{F(\mathbf{x},\mathbf{y}) :=G(\mathbf{x}) + \langle \mathbf{y}, \overline{P} \mathbf{x} \rangle - H(\mathbf{y})\}$, where the functions $G(\cdot)$, $H(\cdot)$, and the the coupling matrix $\overline{P}$ are distributed over a strongly connected network of nodes. GT-GDA is a first-order method that uses gradient tracking to eliminate the dissimilarity caused by heterogeneous data distribution among the nodes. In the most general form, GT-GDA includes a consensus over the local coupling matrices to achieve the optimal (unique) saddle point, however, at the expense of increased communication. To avoid this, we propose a more efficient variant GT-GDA-Lite that does not incur the additional communication and analyze its convergence in various scenarios. We show that GT-GDA converges linearly to the unique saddle point solution when $G(\cdot)$ is smooth and convex, $H(\cdot)$ is smooth and strongly convex, and the global coupling matrix $\overline{P}$ has full column rank. We further characterize the regime under which GT-GDA exhibits a network topology-independent convergence behavior. We next show the linear convergence of GT-GDA to an error around the unique saddle point, which goes to zero when the coupling cost ${\langle \mathbf y, \overline{P} \mathbf x \rangle}$ is common to all nodes, or when $G(\cdot)$ and $H(\cdot)$ are quadratic. Numerical experiments illustrate the convergence properties and importance of GT-GDA and GT-GDA-Lite for several applications.

Variance reduced stochastic optimization over directed graphs with row and column stochastic weights

This paper proposes AB-SAGA, a first-order distributed stochastic optimization method to minimize a finite-sum of smooth and strongly convex functions distributed over an arbitrary directed graph. AB-SAGA removes the uncertainty caused by the stochastic gradients using a node-level variance reduction and subsequently employs network-level gradient tracking to address the data dissimilarity across the nodes. Unlike existing methods that use the nonlinear push-sum correction to cancel the imbalance caused by the directed communication, the consensus updates in AB-SAGA are linear and uses both row and column stochastic weights. We show that for a constant step-size, AB-SAGA converges linearly to the global optimal. We quantify the directed nature of the underlying graph using an explicit directivity constant and characterize the regimes in which AB-SAGA achieves a linear speed-up over its centralized counterpart. Numerical experiments illustrate the convergence of AB-SAGA for strongly convex and nonconvex problems.

Distributed Detection and Mitigation of Biasing Attacks over Multi-Agent Networks

This paper proposes a distributed attack detection and mitigation technique based on distributed estimation over a multi-agent network, where the agents take partial system measurements susceptible to (possible) biasing attacks. In particular, we assume that the system is not locally observable via the measurements in the direct neighborhood of any agent. First, for performance analysis in the attack-free case, we show that the proposed distributed estimation is unbiased with bounded mean-square deviation in steady-state. Then, we propose a residual-based strategy to locally detect possible attacks at agents. In contrast to the deterministic thresholds in the literature assuming an upper bound on the noise support, we define the thresholds on the residuals in a probabilistic sense. After detecting and isolating the attacked agent, a system-digraph-based mitigation strategy is proposed to replace the attacked measurement with a new observationally-equivalent one to recover potential observability loss. We adopt a graph-theoretic method to classify the agents based on their measurements, to distinguish between the agents recovering the system rank-deficiency and the ones recovering output-connectivity of the system digraph. The attack detection/mitigation strategy is specifically described for each type, which is of polynomial-order complexity for large-scale applications. Illustrative simulations support our theoretical results.

Distributed support-vector-machine over dynamic balanced directed networks

In this paper, we consider the binary classification problem via distributed Support-Vector-Machines (SVM), where the idea is to train a network of agents, with limited share of data, to cooperatively learn the SVM classifier for the global database. Agents only share processed information regarding the classifier parameters and the gradient of the local loss functions instead of their raw data. In contrast to the existing work, we propose a continuous-time algorithm that incorporates network topology changes in discrete jumps. This hybrid nature allows us to remove chattering that arises because of the discretization of the underlying CT process. We show that the proposed algorithm converges to the SVM classifier over time-varying weight balanced directed graphs by using arguments from the matrix perturbation theory.

Delay-Tolerant Consensus-based Distributed Estimation: Full-Rank Systems with Potentially Unstable Dynamics

Classical distributed estimation scenarios typically assume timely and reliable exchange of information over the multi-agent network. This paper, in contrast, considers single time-scale distributed estimation of (potentially) unstable full-rank dynamical systems via a multi-agent network subject to transmission time-delays. The proposed networked estimator consists of two steps: (i) consensus on (delayed) a-priori estimates, and (ii) measurement update. The agents only share their a-priori estimates with their in-neighbors over time-delayed transmission links. Considering the most general case, the delays are assumed to be time-varying, arbitrary, unknown, but upper-bounded. In contrast to most recent distributed observers assuming system observability in the neighborhood of each agent, our proposed estimator makes no such assumption. This may significantly reduce the communication/sensing loads on agents in large-scale, while making the (distributed) observability analysis more challenging. Using the notions of augmented matrices and Kronecker product, the geometric convergence of the proposed estimator over strongly-connected networks is proved irrespective of the bound on the time-delay. Simulations are provided to support our theoretical results.

A hybrid variance-reduced method for decentralized stochastic non-convex optimization

This paper considers decentralized stochastic optimization over a network of~$n$ nodes, where each node possesses a smooth non-convex local cost function and the goal of the networked nodes is to find an~$\epsilon$-accurate first-order stationary point of the sum of the local costs. We focus on an online setting, where each node accesses its local cost only by means of a stochastic first-order oracle that returns a noisy version of the exact gradient. In this context, we propose a novel single-loop decentralized hybrid variance-reduced stochastic gradient method, called \texttt{GT-HSGD}, that outperforms the existing approaches in terms of both the oracle complexity and practical implementation. The \texttt{GT-HSGD} algorithm implements specialized local hybrid stochastic gradient estimators that are fused over the network to track the global gradient. Remarkably, \texttt{GT-HSGD} achieves a network-independent oracle complexity of~$O(n^{-1}\epsilon^{-3})$ when the required error tolerance~$\epsilon$ is small enough, leading to a linear speedup with respect to the centralized optimal online variance-reduced approaches that operate on a single node. Numerical experiments are provided to illustrate our main technical results.

A fast randomized incremental gradient method for decentralized non-convex optimization

We study decentralized non-convex finite-sum minimization problems described over a network of nodes, where each node possesses a local batch of data samples. We propose a single-timescale first-order randomized incremental gradient method, termed as GT-SAGA. GT-SAGA is computationally efficient since it evaluates only one component gradient per node per iteration and achieves provably fast and robust performance by leveraging node-level variance reduction and network-level gradient tracking. For general smooth non-convex problems, we show almost sure and mean-squared convergence to a first-order stationary point and describe regimes of practical significance where GT-SAGA achieves a network-independent convergence rate and outperforms the existing approaches respectively. When the global cost function further satisfies the Polyak-Lojaciewisz condition, we show that GT-SAGA exhibits global linear convergence to an optimal solution in expectation and describe regimes of practical interest where the performance is network-independent and improves upon the existing work. Numerical experiments based on real-world datasets are included to highlight the behavior and convergence aspects of the proposed method.

A near-optimal stochastic gradient method for decentralized non-convex finite-sum optimization

This paper describes a $near$-$optimal$ stochastic first-order gradient method for decentralized finite-sum minimization of smooth non-convex functions. Specifically, we propose GT-SARAH that employs a local SARAH-type variance reduction and global gradient tracking to address the stochastic and decentralized nature of the problem. Considering a total number of $N$ cost functions, equally divided over a directed network of $n$ nodes, we show that GT-SARAH finds an $\epsilon$-accurate first-order stationary point in ${\mathcal{O}(N^{1/2}\epsilon^{-1})}$ gradient computations across all nodes, independent of the network topology, when ${n\leq\mathcal{O}(N^{1/2}(1-\lambda)^{3})}$, where ${(1-\lambda)}$ is the spectral gap of the network weight matrix. In this regime, GT-SARAH is thus, to the best our knowledge, the first decentralized method that achieves the algorithmic lower bound for this class of problems. Moreover, GT-SARAH achieves a $non$-$asymptotic$ $linear$ $speedup$, in that, the total number of gradient computations at each node is reduced by a factor of $1/n$ compared to the near-optimal algorithms for this problem class that process all data at a single node. We also establish the convergence rate of GT-SARAH in other regimes, in terms of the relative sizes of the number of nodes $n$, total number of functions $N$, and the network spectral gap $(1-\lambda)$. Over infinite time horizon, we establish the almost sure and mean-squared convergence of GT-SARAH to a first-order stationary point.