arXiv subjects

Anit Kumar Sahu

Publications and source records attributed to Anit Kumar Sahu.

At least 37 records · Page 2Linked to original sources

arXiv2021

You Only Query Once: Effective Black Box Adversarial Attacks with Minimal Repeated Queries

Researchers have repeatedly shown that it is possible to craft adversarial attacks on deep classifiers (small perturbations that significantly change the class label), even in the "black-box" setting where one only has query access to the classifier. However, all prior work in the black-box setting attacks the classifier by repeatedly querying the same image with minor modifications, usually thousands of times or more, making it easy for defenders to detect an ensuing attack. In this work, we instead show that it is possible to craft (universal) adversarial perturbations in the black-box setting by querying a sequence of different images only once. This attack prevents detection from high number of similar queries and produces a perturbation that causes misclassification when applied to any input to the classifier. In experiments, we show that attacks that adhere to this restriction can produce untargeted adversarial perturbations that fool the vast majority of MNIST and CIFAR-10 classifier inputs, as well as in excess of $60-70\%$ of inputs on ImageNet classifiers. In the targeted setting, we exhibit targeted black-box universal attacks on ImageNet classifiers with success rates above $20\%$ when only allowed one query per image, and $66\%$ when allowed two queries per image.

cs.LG↗

arXiv2020

Noisy Batch Active Learning with Deterministic Annealing

We study the problem of training machine learning models incrementally with batches of samples annotated with noisy oracles. We select each batch of samples that are important and also diverse via clustering and importance sampling. More importantly, we incorporate model uncertainty into the sampling probability to compensate for poor estimation of the importance scores when the training data is too small to build a meaningful model. Experiments on benchmark image classification datasets (MNIST, SVHN, CIFAR10, and EMNIST) show improvement over existing active learning strategies. We introduce an extra denoising layer to deep networks to make active learning robust to label noises and show significant improvements.

cs.LG↗

arXiv2020

Gaussian MRF Covariance Modeling for Efficient Black-Box Adversarial Attacks

We study the problem of generating adversarial examples in a black-box setting, where we only have access to a zeroth order oracle, providing us with loss function evaluations. Although this setting has been investigated in previous work, most past approaches using zeroth order optimization implicitly assume that the gradients of the loss function with respect to the input images are \emph{unstructured}. In this work, we show that in fact substantial correlations exist within these gradients, and we propose to capture these correlations via a Gaussian Markov random field (GMRF). Given the intractability of the explicit covariance structure of the MRF, we show that the covariance structure can be efficiently represented using the Fast Fourier Transform (FFT), along with low-rank updates to perform exact posterior estimation under this model. We use this modeling technique to find fast one-step adversarial attacks, akin to a black-box version of the Fast Gradient Sign Method~(FGSM), and show that the method uses fewer queries and achieves higher attack success rates than the current state of the art. We also highlight the general applicability of this gradient modeling setup.

cs.LG↗

arXiv2020

Data-driven Thermal Model Inference with ARMAX, in Smart Environments, based on Normalized Mutual Information

Understanding the models that characterize the thermal dynamics in a smart building is important for the comfort of its occupants and for its energy optimization. A significant amount of research has attempted to utilize thermodynamics (physical) models for smart building control, but these approaches remain challenging due to the stochastic nature of the intermittent environmental disturbances. This paper presents a novel data-driven approach for indoor thermal model inference, which combines an Autoregressive Moving Average with eXogenous inputs model (ARMAX) with a Normalized Mutual Information scheme (NMI). Based on this information-theoretic method, NMI, causal dependencies between the indoor temperature and exogenous inputs are explicitly obtained as a guideline for the ARMAX model to find the dominating inputs. For validation, we use three datasets based on building energy systems-against which we compare our method to an autoregressive model with exogenous inputs (ARX), a regularized ARMAX model, and state-space models.

eess.SY↗

arXiv2020

Federated Optimization in Heterogeneous Networks

Federated Learning is a distributed learning paradigm with two key challenges that differentiate it from traditional distributed optimization: (1) significant variability in terms of the systems characteristics on each device in the network (systems heterogeneity), and (2) non-identically distributed data across the network (statistical heterogeneity). In this work, we introduce a framework, FedProx, to tackle heterogeneity in federated networks. FedProx can be viewed as a generalization and re-parametrization of FedAvg, the current state-of-the-art method for federated learning. While this re-parameterization makes only minor modifications to the method itself, these modifications have important ramifications both in theory and in practice. Theoretically, we provide convergence guarantees for our framework when learning over data from non-identical distributions (statistical heterogeneity), and while adhering to device-level systems constraints by allowing each participating device to perform a variable amount of work (systems heterogeneity). Practically, we demonstrate that FedProx allows for more robust convergence than FedAvg across a suite of realistic federated datasets. In particular, in highly heterogeneous settings, FedProx demonstrates significantly more stable and accurate convergence behavior relative to FedAvg---improving absolute test accuracy by 22% on average.

cs.LG↗

arXiv2020

FedDANE: A Federated Newton-Type Method

Federated learning aims to jointly learn statistical models over massively distributed remote devices. In this work, we propose FedDANE, an optimization method that we adapt from DANE, a method for classical distributed optimization, to handle the practical constraints of federated learning. We provide convergence guarantees for this method when learning over both convex and non-convex functions. Despite encouraging theoretical results, we find that the method has underwhelming performance empirically. In particular, through empirical simulations on both synthetic and real-world datasets, FedDANE consistently underperforms baselines of FedAvg and FedProx in realistic federated settings. We identify low device participation and statistical device heterogeneity as two underlying causes of this underwhelming performance, and conclude by suggesting several directions of future work.

cs.LG↗

arXiv2019

MATCHA: Speeding Up Decentralized SGD via Matching Decomposition Sampling

This paper studies the problem of error-runtime trade-off, typically encountered in decentralized training based on stochastic gradient descent (SGD) using a given network. While a denser (sparser) network topology results in faster (slower) error convergence in terms of iterations, it incurs more (less) communication time/delay per iteration. In this paper, we propose MATCHA, an algorithm that can achieve a win-win in this error-runtime trade-off for any arbitrary network topology. The main idea of MATCHA is to parallelize inter-node communication by decomposing the topology into matchings. To preserve fast error convergence speed, it identifies and communicates more frequently over critical links, and saves communication time by using other links less frequently. Experiments on a suite of datasets and deep neural networks validate the theoretical analyses and demonstrate that MATCHA takes up to $5\times$ less time than vanilla decentralized SGD to reach the same training loss.

cs.LG↗

arXiv2019

Black-box Adversarial Attacks with Bayesian Optimization

We focus on the problem of black-box adversarial attacks, where the aim is to generate adversarial examples using information limited to loss function evaluations of input-output pairs. We use Bayesian optimization~(BO) to specifically cater to scenarios involving low query budgets to develop query efficient adversarial attacks. We alleviate the issues surrounding BO in regards to optimizing high dimensional deep learning models by effective dimension upsampling techniques. Our proposed approach achieves performance comparable to the state of the art black-box adversarial attacks albeit with a much lower average query count. In particular, in low query budget regimes, our proposed method reduces the query count up to $80\%$ with respect to the state of the art methods.

cs.LG↗

arXiv2019

Federated Learning: Challenges, Methods, and Future Directions

Federated learning involves training statistical models over remote devices or siloed data centers, such as mobile phones or hospitals, while keeping data localized. Training in heterogeneous and potentially massive networks introduces novel challenges that require a fundamental departure from standard approaches for large-scale machine learning, distributed optimization, and privacy-preserving data analysis. In this article, we discuss the unique characteristics and challenges of federated learning, provide a broad overview of current approaches, and outline several directions of future work that are relevant to a wide range of research communities.

cs.LG↗

arXiv2019

Distributed stochastic optimization with gradient tracking over strongly-connected networks

In this paper, we study distributed stochastic optimization to minimize a sum of smooth and strongly-convex local cost functions over a network of agents, communicating over a strongly-connected graph. Assuming that each agent has access to a stochastic first-order oracle ($\mathcal{SFO}$), we propose a novel distributed method, called $\mathcal{S}$-$\mathcal{AB}$, where each agent uses an auxiliary variable to asymptotically track the gradient of the global cost in expectation. The $\mathcal{S}$-$\mathcal{AB}$ algorithm employs row- and column-stochastic weights simultaneously to ensure both consensus and optimality. Since doubly-stochastic weights are not used, $\mathcal{S}$-$\mathcal{AB}$ is applicable to arbitrary strongly-connected graphs. We show that under a sufficiently small constant step-size, $\mathcal{S}$-$\mathcal{AB}$ converges linearly (in expected mean-square sense) to a neighborhood of the global minimizer. We present numerical simulations based on real-world data sets to illustrate the theoretical results.

cs.LG↗

arXiv2019

Towards Gradient Free and Projection Free Stochastic Optimization

This paper focuses on the problem of \emph{constrained} \emph{stochastic} optimization. A zeroth order Frank-Wolfe algorithm is proposed, which in addition to the projection-free nature of the vanilla Frank-Wolfe algorithm makes it gradient free. Under convexity and smoothness assumption, we show that the proposed algorithm converges to the optimal objective function at a rate $O\left(1/T^{1/3}\right)$, where $T$ denotes the iteration count. In particular, the primal sub-optimality gap is shown to have a dimension dependence of $O\left(d^{1/3}\right)$, which is the best known dimension dependence among all zeroth order optimization algorithms with one directional derivative per iteration. For non-convex functions, we obtain the \emph{Frank-Wolfe} gap to be $O\left(d^{1/3}T^{-1/4}\right)$. Experiments on black-box optimization setups demonstrate the efficacy of the proposed algorithm.

math.OC↗

arXiv2018

Managing App Install Ad Campaigns in RTB: A Q-Learning Approach

Real time bidding (RTB) enables demand side platforms (bidders) to scale ad campaigns across multiple publishers affiliated to an RTB ad exchange. While driving multiple campaigns for mobile app install ads via RTB, the bidder typically has to: (i) maintain each campaign's efficiency (i.e., meet advertiser's target cost-per-install), (ii) be sensitive to advertiser's budget, and (iii) make profit after payouts to the ad exchange. In this process, there is a sense of delayed rewards for the bidder's actions; the exchange charges the bidder right after the ad is shown, but the bidder gets to know about resultant installs after considerable delay. This makes it challenging for the bidder to decide beforehand the bid (and corresponding cost charged to advertiser) for each ad display opportunity. To jointly handle the objectives mentioned above, we propose a state space based policy which decides the exchange bid and advertiser cost for each opportunity. The state space captures the current efficiency, budget utilization and profit. The policy based on this state space is trained on past decisions and outcomes via a novel Q-learning algorithm which accounts for the delay in install notifications. In our experiments based on data from app install campaigns managed by Yahoo's Gemini advertising platform, the Q-learning based policy led to a significant increase in the profit and number of efficient campaigns.

cs.GT↗

arXiv2018

Communication-Efficient Distributed Strongly Convex Stochastic Optimization: Non-Asymptotic Rates

We examine fundamental tradeoffs in iterative distributed zeroth and first order stochastic optimization in multi-agent networks in terms of \emph{communication cost} (number of per-node transmissions) and \emph{computational cost}, measured by the number of per-node noisy function (respectively, gradient) evaluations with zeroth order (respectively, first order) methods. Specifically, we develop novel distributed stochastic optimization methods for zeroth and first order strongly convex optimization by utilizing a probabilistic inter-agent communication protocol that increasingly sparsifies communications among agents as time progresses. Under standard assumptions on the cost functions and the noise statistics, we establish with the proposed method the $O(1/(C_{\mathrm{comm}})^{4/3-ζ})$ and $O(1/(C_{\mathrm{comm}})^{8/9-ζ})$ mean square error convergence rates, for the first and zeroth order optimization, respectively, where $C_{\mathrm{comm}}$ is the expected number of network communications and $ζ>0$ is arbitrarily small. The methods are shown to achieve order-optimal convergence rates in terms of computational cost~$C_{\mathrm{comp}}$, $O(1/C_{\mathrm{comp}})$ (first order optimization) and $O(1/(C_{\mathrm{comp}})^{2/3})$ (zeroth order optimization), while achieving the order-optimal convergence rates in terms of iterations. Experiments on real-life datasets illustrate the efficacy of the proposed algorithms.

math.OC↗

arXiv2018

$\mathcal{CIRFE}$: A Distributed Random Fields Estimator

This paper presents a communication efficient distributed algorithm, $\mathcal{CIRFE}$ of the \emph{consensus}+\emph{innovations} type, to estimate a high-dimensional parameter in a multi-agent network, in which each agent is interested in reconstructing only a few components of the parameter. This problem arises for example when monitoring the high-dimensional distributed state of a large-scale infrastructure with a network of limited capability sensors and where each sensor is tasked with estimating some local components of the state. At each observation sampling epoch, each agent updates its local estimate of the parameter components in its interest set by simultaneously processing the latest locally sensed information~(\emph{innovations}) and the parameter estimates from agents~(\emph{consensus}) in its communication neighborhood given by a time-varying possibly sparse graph. Under minimal conditions on the inter-agent communication network and the sensing models, almost sure convergence of the estimate sequence at each agent to the components of the true parameter in its interest set is established. Furthermore, the paper establishes the performance of $\mathcal{CIRFE}$ in terms of asymptotic covariance of the estimate sequences and specifically characterizes the dependencies of the component wise asymptotic covariance in terms of the number of agents tasked with estimating it. Finally, simulation experiments demonstrate the efficacy of $\mathcal{CIRFE}$.

math.OC↗

arXiv2018

Convergence rates for distributed stochastic optimization over random networks

We establish the O($\frac{1}{k}$) convergence rate for distributed stochastic gradient methods that operate over strongly convex costs and random networks. The considered class of methods is standard each node performs a weighted average of its own and its neighbors solution estimates (consensus), and takes a negative step with respect to a noisy version of its local functions gradient (innovation). The underlying communication network is modeled through a sequence of temporally independent identically distributed (i.i.d.) Laplacian matrices connected on average, while the local gradient noises are also i.i.d. in time, have finite second moment, and possibly unbounded support. We show that, after a careful setting of the consensus and innovations potentials (weights), the distributed stochastic gradient method achieves a (order-optimal) O($\frac{1}{k}$) convergence rate in the mean square distance from the solution. This is the first order-optimal convergence rate result on distributed strongly convex stochastic optimization when the network is random and/or the gradient noises have unbounded support. Simulation examples confirm the theoretical findings.

math.OC↗

arXiv2018

Distributed Zeroth Order Optimization Over Random Networks: A Kiefer-Wolfowitz Stochastic Approximation Approach

We study a standard distributed optimization framework where $N$ networked nodes collaboratively minimize the sum of their local convex costs. The main body of existing work considers the described problem when the underling network is either static or deterministically varying, and the distributed optimization algorithm is of first or second order, i.e., it involves the local costs' gradients and possibly the local Hessians. In this paper, we consider the currently understudied but highly relevant scenarios when: 1) only noisy function values' estimates are available (no gradients nor Hessians can be evaluated); and 2) the underlying network is randomly varying (according to an independent, identically distributed process). For the described random networks-zeroth order optimization setting, we develop a distributed stochastic approximation method of the Kiefer-Wolfowitz type. Furthermore, under standard smoothness and strong convexity assumptions on the local costs, we establish the $O(1/k^{1/2})$ mean square convergence rate for the method -- the rate that matches that of the method's centralized counterpart under equivalent conditions.

math.OC↗

arXiv2018

Communication Optimality Trade-offs For Distributed Estimation

This paper proposes $\mathbf{C}$ommunication efficient $\mathbf{RE}$cursive $\mathbf{D}$istributed estimati$\mathbf{O}$n algorithm, $\mathcal{CREDO}$, for networked multi-worker setups without a central master node. $\mathcal{CREDO}$ is designed for scenarios in which the worker nodes aim to collaboratively estimate a vector parameter of interest using distributed online time-series data at the individual worker nodes. The individual worker nodes iteratively update their estimate of the parameter by assimilating latest locally sensed information and estimates from neighboring worker nodes exchanged over a (possibly sparse) time-varying communication graph. The underlying inter-worker communication protocol is adaptive, making communications increasingly (probabilistically) sparse as time progresses. Under minimal conditions on the inter-worker information exchange network and the sensing models, almost sure convergence of the estimate sequences at the worker nodes to the true parameter is established. Further, the paper characterizes the performance of $\mathcal{CREDO}$ in terms of asymptotic covariance of the estimate sequences and specifically establishes the achievability of optimal asymptotic covariance. The analysis reveals an interesting interplay between the algorithm's communication cost~$\mathcal{C}_{t}$ (over $t$ time-steps) and the asymptotic covariance. Most notably, it is shown that $\mathcal{CREDO}$ may be designed to achieve a $Θ\left(\mathcal{C}_{t}^{-2+ζ}\right)$ decay of the mean square error~($ζ>0$, arbitrarily small) at each worker node, which significantly improves over the existing $ Θ\left(\mathcal{C}_{t}^{-1}\right)$ rates. Simulation examples on both synthetic and real data sets demonstrate $\mathcal{CREDO}$'s communication efficiency.

math.OC↗

arXiv2017

Recursive Distributed Detection for Composite Hypothesis Testing: Nonlinear Observation Models in Additive Gaussian Noise

This paper studies recursive composite hypothesis testing in a network of sparsely connected agents. The network objective is to test a simple null hypothesis against a composite alternative concerning the state of the field, modeled as a vector of (continuous) unknown parameters determining the parametric family of probability measures induced on the agents' observation spaces under the hypotheses. Specifically, under the alternative hypothesis, each agent sequentially observes an independent and identically distributed time-series consisting of a (nonlinear) function of the true but unknown parameter corrupted by Gaussian noise, whereas, under the null, they obtain noise only. Two distributed recursive generalized likelihood ratio test type algorithms of the \emph{consensus+innovations} form are proposed, namely $\mathcal{CIGLRT-L}$ and $\mathcal{CIGLRT-NL}$, in which the agents estimate the underlying parameter and in parallel also update their test decision statistics by simultaneously processing the latest local sensed information and information obtained from neighboring agents. For $\mathcal{CIGLRT-NL}$, for a broad class of nonlinear observation models and under a global observability condition, algorithm parameters which ensure asymptotically decaying probabilities of errors~(probability of miss and probability of false detection) are characterized. For $\mathcal{CIGLRT-L}$, a linear observation model is considered and upper bounds on large deviations decay exponent for the error probabilities are obtained.

cs.IT↗