A
Anit Kumar Sahu
Researcher at Bosch
Publications - 67
Citations - 5305
Anit Kumar Sahu is an academic researcher from Bosch. The author has contributed to research in topics: Computer science & Independent and identically distributed random variables. The author has an hindex of 17, co-authored 55 publications receiving 2353 citations. Previous affiliations of Anit Kumar Sahu include Carnegie Mellon University & Tufts University.
Papers
More filters
Proceedings ArticleDOI
Data-driven Thermal Model Inference with ARMAX, in Smart Environments, based on Normalized Mutual Information
Zhanhong Jiang,Jonathan Francis,Anit Kumar Sahu,Sirajum Munir,Charles Shelton,Anthony Rowe,Mario Berges +6 more
TL;DR: A novel data-driven approach for indoor thermal model inference is presented, which combines an Autoregressive Moving Average with eXogenous inputs model (ARMAX) with a Normalized Mutual Information scheme (NMI).
Proceedings ArticleDOI
Simple and Efficient Hard Label Black-box Adversarial Attacks in Low Query Budget Regimes
TL;DR: In this article, a simple and efficient Bayesian Optimization (BO) based approach for developing black-box adversarial attacks is proposed. But the method is limited to output label (hard-label) to a queried data input.
Posted Content
Recursive Distributed Detection for Composite Hypothesis Testing: Algorithms and Asymptotics
Anit Kumar Sahu,Soummya Kar +1 more
TL;DR: Two distributed recursive generalized likelihood ratio test type algorithms of the \emph{consensus+innovations} form are proposed, namely $\mathcal{CILRT}$ and $\mathCal{CIGLRT}$, 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.
Proceedings ArticleDOI
Non-asymptotic rates for communication efficient distributed zeroth order strongly convex optimization
TL;DR: Distributed stochastic optimization methods for zeroth order strongly convex optimization that are based on an adaptive probabilistic sparsifying communications protocol are developed and established with the proposed method O(1/(Ccomm)2/3-ζ) mean square error (MSE) convergence rates.
Journal ArticleDOI
Communication efficient distributed weighted non-linear least squares estimation
TL;DR: It is rigorously proved that CREDO−Nℒ$\mathcal {CREDO-NL}$ achieves significantly faster mean squared error rates in terms of the elapsed communication cost over existing alternatives, and the considered simulation experiments show communication savings by at least an order of magnitude.