My current research interest broadly lies in Applied Probability with a specific focus on Statistical Methods and Monte Carlo Simulation algorithms. In particular, I am interested in Machine Learning, Spatial Point Processes, Perfect Sampling, Large Deviations Theory, and Variance Reduction techniques.
Research (including the manuscripts in preparation):
1. Unbiased Estimation of the Reciprocal Mean for Non-negative Random Variables S. B. Moka, D. P. Kroese, and S. Juneja
(arXiv) (Accepted for WSC 2019)
2. Towards Q-learning the Whittle Index for Restless Bandits Jing Fu, Yoni Nazarathy, Sarat Moka, Peter Taylor.
(Link) (Accepted for Australian & New Zealand Control Conference 2019)
3. Perfect Sampling for Gibbs Point Processes using Partial Rejection Sampling S. B. Moka and D. P. Kroese
(arXiv) (Under review in Bernoulli)
4. Chromosome arm aneuploidies shape tumour evolution, cancer prognosis and precision oncology
Thu H. M. Nguyen, Ankit Shukla, S. B. Moka, Jonathan Ellis, John Grady, Harald Oey, Alexandre S. Cristino, Kum Kum Khanna, Dirk P. Kroese, Lutz Krause, Eloise Dray, J. Lynn Fink, Pascal H. G. Duijf.
(Under review in Nature Communications)
5. Diffusion Parameters of Flows in Stable Queueing Networks S. B. Moka, Y. Nazarathy, and W. Scheinhardt
(Under review in Stochastic Systems)
6. Analysis of Perfect Sampling Methods for Hard-sphere Models S. B. Moka, S. Juneja, and M. R. H. Mandjes
ACM SIGMETRICS Performance Evaluation Review 45 (2), 69-75, 2018
7. Rejection and Importance Sampling based Perfect Simulation for Gibbs Processes with a focus on Hard-sphere Models
S. B. Moka, S. Juneja, and M. R. H. Mandjes
(Under review in Advances in Applied Probability) (arXiv preprint)
8. Spatial loss systems: Exact simulation and rare event behavior S. Foss, S. Juneja, M. R. H. Mandjes, and S. B. Moka
ACM SIGMETRICS Performance Evaluation Review 43 (2), 3-6, 2015
9. Regenerative simulation for queueing networks with exponential or heavier tail arrival distributions S. B. Moka and S. Juneja
ACM Transactions on Modeling and Computer Simulation (TOMACS) 25 (4), 22, 2015
10. Regenerative simulation for multiclass open queueing networks S. B. Moka and S. Juneja
Winter Simulation Conference (WSC), 2013 Winter, 643-654, 2013
Large Deviation Theory
Spatial Point Processes
PhD in System Science from Tata Institute of Fundamental Research, Mumbai, India
ME in Telecommunication from Indian Institute of Science, Bangaluru, India
BE in Electronics and Communication Engineering, Andhra University, India
Rare events such as the state-wide power blackout in South Australia in September 2016, natural disasters such as floods and bushfires, or the ensuing chaos when parts of a complex interconnected systems such as the internet fail, are difficult for researchers to simulate or model. They're the primary interest of ACEMS Chief Investigator Dirk Kroese.
Prizes, awards and special recognition
ACEMS Sampling and Exploration Competition, 2017 RESEARCH FELLOW CATEGORY - First Prize was awarded to Sarat Babu Moka. Awarded from the ACEMS.