I am a postdoctoral researcher working with Michael Jordan and Martin Wainwright at UC Berkeley, in the departments of Statistics & EECS, affiliated with the RiseLab/AmpLab and BLISS/WiFo. I recently earned my PhD thesis under Larry Wasserman and Aarti Singh at Carnegie Mellon University, in the Departments of Statistics & Machine Learning, where my thesis on Computational & Statistical Advances in Testing & Learning won the Umesh K. Gavaskar Memorial Thesis Award in Statistics. Earlier, I completed my Bachelors thesis under Supratik Chakraborty at IIT Bombay from the Department of CSE.

Papers (by topic)

False Discovery Rate control (in structured, dynamic or interactive settings)


A Unified Treatment of Multiple Testing with Prior Knowledge
A. Ramdas, R. Barber, M. Wainwright, M. Jordan [Arxiv]
(in submission, Annals of Statistics)
The p-filter: multi-layer FDR control for grouped hypotheses
R. F. Barber*, A. Ramdas* [Arxiv] [code] [JRSSB]
(JRSSB) Journal of the Royal Statistical Society -- Series B (Methodology), 2016
Optimal rates and tradeoffs for multiple testing
M. Rabinovich, A. Ramdas, M. Wainwright, M. Jordan [arxiv]
(in submission, the Annals of Statistics)
Multi (A)rmed/(B)andit testing with online FDR control
F. Yang, A. Ramdas, K. Jamieson, M. Wainwright [arxiv]
(in submission, Neural Information Processing Systems)
Online FDR control with decaying memory and weights
A. Ramdas, F. Yang, M. Wainwright, M. Jordan [pre]
(in submission, Neural Information Processing Systems)
Decentralized FDR control on sensor networks
A. Ramdas, J. Chen, M. Wainwright, M. Jordan [pre]
(CDC) IEEE Conference on Decision and Control, 2017
Sequential FDR control for directed acyclic graphs
A. Ramdas, J. Chen, M. Wainwright, M. Jordan [pre]
(in preparation)
Interactive accumulation tests for structured multiple testing
L. Lei, A. Ramdas, W. Fithian [pre]
(in preparation)


Hypothesis testing (in nonparametric, structured or high-dimensional settings)


Generative models and model criticism via optimized Maximum Mean Discrepancy
D. Sutherland, H. Tung, H. Strathmann, S. De, A. Ramdas, A. Smola, A. Gretton [Arxiv][ICLR]
(ICLR) 5th International Conference on Learning Representations, Toulon, 2017
Adaptivity & computation-statistics tradeoffs for kernel & distance based high-dimensional two sample testing
A. Ramdas, S. Reddi, B. Poczos, A. Singh, L. Wasserman [Arxiv]
(in revision)
Minimax lower bounds for linear independence testing
D. Isenberg*, A. Ramdas*, A. Singh, L. Wasserman [Arxiv][ISIT]
(ISIT) IEEE International Symposium on Information Theory, Barcelona, 2016
Fast two-sample testing with analytic representations of probability measures
K. Chwialkowski, A. Ramdas, D. Sejdinovic, A. Gretton [Arxiv] [github] [NIPS]
(NIPS) 29th Conference on Neural Information Processing Systems, Montreal, 2015
Nonparametric independence testing for small sample sizes
A. Ramdas*, L. Wehbe* [Arxiv] [IJCAI]
(IJCAI) 24th International Joint Conference on Artificial Intelligence, Buenos Aires, 2015
On the high-dimensional power of a linear-time two sample test under mean-shift alternatives
S. Reddi*, A. Ramdas*, A. Singh, B. Poczos, L. Wasserman [AISTATS] [Arxiv] [pdf] [supp]
(AISTATS) 18th International Conference on Artificial Intelligence and Statistics, San Diego, 2015
On the decreasing power of kernel and distance based nonparametric hypothesis tests in high dimensions
A. Ramdas*, S. Reddi*, B. Poczos, A. Singh, L. Wasserman [AAAI] [Arxiv] [pdf] [supp]
(AAAI) 29th AAAI Conference on Artifical Intelligence, Austin, 2015
Classification accuracy as a proxy for two sample testing
A. Ramdas, A. Singh, L. Wasserman [Arxiv]
(in revision)
On kernel methods for covariates that are rankings
H. Mania, A. Ramdas, M. Wainwright, M. Jordan, B. Recht [Arxiv]
(in submission, Journal of American Statistical Association)
On Wasserstein two sample testing and related families of nonparametric tests
A. Ramdas*, N. Garcia*, M. Cuturi [Arxiv] [Entropy]
(Ent) Entropy, Special Issue on Statistical Significance and the Logic of Hypothesis Testing, 2017


Convex optimization


Iterative methods for solving factorized linear systems
A. Ma, D. Needell, A. Ramdas [Arxiv]
(in submission, SIAM Journal on Matrix Analysis)
Rows vs columns : randomized Kaczmarz or Gauss-Seidel for ridge regression
A. Hefny*, D. Needell*, A. Ramdas*, [Arxiv]
(SISC) SIAM Journal on Scientific Computing, 2017
Towards a deeper geometric, analytic and algorithmic understanding of margins
A. Ramdas, J. Pena [Arxiv] [OMS]
(OMS) Optimization Methods and Software, 2015
Convergence properties of the randomized extended Gauss-Seidel and Kaczmarz methods
A. Ma*, D. Needell*, A. Ramdas* [Arxiv] [SIMAX]
(SIMAX) SIAM Journal on Matrix Analysis and Applications, 2015
Fast & flexible ADMM algorithms for trend filtering
A. Ramdas*, R. Tibshirani* [Arxiv] [JCGS] [github `glmgen'] [50 min. talk]
(JCGS) Journal of Computational and Graphical Statistics, 2015
Margins, kernels and non-linear smoothed perceptrons
A. Ramdas, J. Pena [Arxiv] [ICML] [pdf] [supp] [20-min oral]
(ICML) 31st International Conference on Machine Learning, Beijing, 2014
Optimal rates for stochastic convex optimization under Tsybakov's noise condition
A. Ramdas, A. Singh [ICML] [Arxiv] [pdf] [supp] [20-min oral]
(ICML) 30th International Conference on Machine Learning, Atlanta, 2013


Other problems of a sequential nature


Online generation of low-discrepancy sequences: from Monte-Carlo to Quasi-Monte-Carlo by retrying
O. N. Feldheim*, A. Ramdas*, R. Dwivedi* [pre]
(in submission, Foundations of Computer Science)
Function-specific mixing times and concentration away from equilibrium
M. Rabinovich, A. Ramdas, M. Wainwright, M. Jordan [Arxiv]
(in submission, Bernoulli)
Sequential nonparametric testing with the law of the iterated logarithm
A. Balsubramani*, A. Ramdas* [Arxiv][UAI]
(UAI) 32nd Conference on Uncertainty in Artificial Intelligence, New York, 2016
An analysis of active learning with uniform feature noise
A. Ramdas, A. Singh, L. Wasserman, B. Poczos [Arxiv] [AISTATS] [pdf] [supp] [25-min oral]
(AISTATS) 17th International Conference on Artificial Intelligence and Statistics, Reykjavik, 2014
Algorithmic connections between active learning and stochastic convex optimization
A. Ramdas, A. Singh [Arxiv] [ALT] [pdf] [25-min oral]
(ALT) 24th International Conference on Algorithmic Learning Theory, Singapore, 2013


Other collaborations


Decoding from pooled data (II): sharp information-theoretic bounds
A. El-Alaoui, A. Ramdas, F. Krzakala, L. Zdeborova, M. Jordan [Arxiv]
(in submission, Annals of Applied Probability)
Decoding from pooled data (I): phase transitions of message passing
A. El-Alaoui, A. Ramdas, F. Krzakala, L. Zdeborova, M. Jordan [Arxiv]
(ISIT) IEEE International Symposium on Information Theory, Aachen, 2017
Asymptotic behavior of Lq-based Laplacian regularization in semi-supervised learning
A. El-Alaoui, X. Cheng, A. Ramdas, M. Wainwright, M. Jordan [Arxiv][COLT]
(COLT) 29th International Conference on Learning Theory, New York, 2016
Regularized brain reading with shrinkage and smoothing
L. Wehbe, A. Ramdas, R. Steorts, C. Shalizi [Arxiv] [AoAS]
(AoAS) Annals of Applied Statistics, 2015
Simultaneously uncovering the patterns of brain regions involved in different story reading subprocesses
L. Wehbe, B. Murphy, P. Talukdar, A. Fyshe, A. Ramdas, T. Mitchell [website] [PLOS] [pdf] [supp]
(PLoS ONE) Public Library of Science ONE, 2014


Talks

Universities

Modern advances and open directions in multiple testing (UT Southwestern, Medical Center, '17)
Multi (A)rmed/(B)andit testing with online FDR control (UT Austin, EE, '17)
Multi (A)rmed/(B)andit testing with online FDR control (Stanford, MS&E, '17)
Multi (A)rmed/(B)andit testing with online FDR control (CMU, ML/AI, '17)
Decentralized testing on networks with FDR control (UC Berkeley, EE, '17)
False Discovery Rate - a tutorial and new directions (IIT Bombay, EE, '16)
p-Filter: FDR control for grouped hypotheses (Wharton, Stat., '16)
p-Filter: FDR control for grouped hypotheses (Stanford, Stat., '16)
p-Filter: FDR control for grouped hypotheses (UC Davis, Stat., '16)
p-Filter: FDR control for grouped hypotheses (UCBerkeley AmpLab, CS, '16)
Asymptotics of Laplacian regularization in semi-supervised learning (CMU, ML, '16)
Margins - algorithms, geometry and convex analysis (CMU, CS, '15)
Adaptivity in high-dimensional two sample testing (CMU, ML, '15)
Adaptivity in high-dimensional two sample testing (UC Berkeley, CS, '14)
Adaptivity in high-dimensional two sample testing (Kyoto University, Stat., '14)
Adaptivity in high-dimensional two sample testing (ISM Tachikawa, Stat., '14)
Fast & flexible algorithms for trend filtering (Gatsby, Neuro., '14)
Fast & flexible algorithms for trend filtering (CMU, ML, '14)
Connecting active learning and stochastic optimization (CMI Chennai, Math., '13)
Connecting active learning and stochastic optimization (IIT Madras, CS, '13)
Connecting active learning and stochastic optimization (CMU, ML, '13)

Industry

Multi (A)rmed/(B)andit testing with online FDR control (Uber, San Francisco, '17)
Sequential and multiple hypothesis testing in modern ML (Groupon, Palo Alto, '16)
p-Filter: FDR control for grouped hypotheses (Lawrence National Labs, Livermore, '16)
Sequential nonparametric testing (Alibaba, Seattle, '15)
Sequential nonparametric testing (Google, Pittsburgh, '15)
Fast & flexible algorithms for trend filtering (Microsoft Research, Cambridge, '14)
Active learning & stochastic optimization (IBM Research, Bangalore, '13)
Connecting statistical & logical inference (Microsoft Research, Cambridge, '12)

Conferences

On kernel methods for covariates that are rankings (CNA, '17)
A unified treatment of multiple testing with prior knowledge (MCP, '17)
Optimal rates and tradeoffs in multiple testing (ICSA, '17)
Optimal rates and tradeoffs in multiple testing (WHOA-PSI2, '17)
Sequential nonparametric testing using the law of the iterated logarithm (ITA, '17)
Online gen. of low-discrepancy sequences: Monte-Carlo to QMC by retrying (MCM, '17)
A unified framework for multiple testing with prior knowledge (NIPS WADAPT, '16)
Sequential nonparametric testing using the law of the iterated logarithm (Lorentz, '16)
Function-specific mixing times and concentration away from equilibrium (MCQMC, '16)
Minimax bounds for linear independence testing (ISIT, '16)
Function-specific mixing times and concentration away from equlibrium (ISBA, '16)
Beyond worst-case mixing times for markov chains (ITA, '16)
Nonparametric independence testing for small sample sizes (IJCAI, '15)
Adaptivity in high-dimensional two-sample testing (JSM, '15)
Sequential nonparametric testing using the law of the iterated logarithm (IWSM, '15)
Margins, kernels and nonlinear smoothed perceptrons (ICML, '14)
Active learning with uniform feature noise (AISTATS, '14)
Connecting convex optimization and active learning (NIPS OPT, '13)
Algorithmic connections between convex optimization and active learning (ALT, '13)
Optimal convex optimization under Tsybakov noise condition (ICML, '13)

Life outside work

Some current pursuits involve getting better at yoga and cooking, living a plastic-free life, and fundraising as a training ride leader for the AIDS LifeCycle 2017.

Endurance Sports

  • Triathlons --- I completed a full Ironman 140.6 (2.4 mile swim, 112 mile bike, 26.2 mile run) in Louisville, Kentucky (August 25, 2013). I also completed a half Ironman (Grafham, UK, Summer '12) and several Olympic distance triathlons (Marlow, Pittsburgh, Tahoe).
  • Marathons --- I've run a few marathons (Pittsburgh thrice, Columbus) and many half-marathons (Chicago, Atlanta, Philadelphia, Wales, Birmingham, Berkeley) and many 15mi/20mi/30K races around Pittsburgh (Spring Thaw, Just a Short Run, Marshall Mangler).
  • Biking --- I recently completed the fantastic 2016 AIDS LifeCycle ride, a 550 mile ride from SF to LA to raise money for AIDS and LGBT causes. Another memorable ride was the MS-150, a 150-mile weekend bike ride through the hills and farms of central Pennsylvania to raise money for Multiple Sclerosis patients. I also did the amazing 500km ride from Pittsburgh to Washington DC, which is a fully off-road trail ride through parks and forests and rivers and canals (Great Allegheny Passage, Chesapeake-Ohio Canal).

Adventure Sports

  • Land --- I completed a fantastic 26-day Basic Mountaineering Course in the Himalayas (July 1-26, 2006), introducing us to climbing, jumaring, knots, rappelling, river-crossing, snowcraft, icecraft, safety, rescue, survival, etc where we reached a maximum height of 18,200 feet. I also finished a backpacking school in 2013 with the Explorer's Club of Pittsburgh (ECP) that covered clothing, cooking, weather, first aid, planning, leading, etc and some wonderful overnight hikes. The ECP's great rock-climbing school was also very memorable, as it taught us a lot about the theory and safe practice of outdoor top-roping, using high quality gear to climb spectacular routes, peaking with our ascent of the iconic Seneca Rocks. I completed a certified Wilderness First Aid course in 2012. I also enjoy indoor bouldering, and have done orienteering in the past.
  • Air --- I completed an Advanced Free Fall (part A) skydiving course that ended with my first solo jump and landing. I've also been paragliding and parasailing, but I'd like to do a course in those too.
  • Water --- I completed the PADI open water certification for scuba diving, and have had the pleasure of diving around the world (France, Iceland, Singapore, Florida, etc).

Athletic Sports

  • Team - Cricket was my main obsession growing up - I played for the Oman U-13 team in the Gulf Cup (runners-up, Kuwait, 2002) and the Oman U-15 team in the Asia Cup (5th out of 14, UAE, 2004). I also played for the IIT Bombay team in three Inter-IIT tournaments (silver medallists, 2005) and around Bombay and the CMU team in a US-wide tournament. I used to play for my high school teams as well, in Oman and India. In IITB, I also played and won several sports for my hostel including basketball, water polo, football, hockey, etc and won several hostel Sports Colors and passing out Sports Color for these.
  • Individual - In IITB, I also represented my hostel's winning teams in sports like badminton and table tennis. At CMU, I picked up squash, racketball and tennis.

Cultural Activities

  • International - I received the Bronze and Silver award in the International Award for Young People (Duke of Edinburgh Award Scheme). I also represented India at a 10-day cultural exchange program in Singapore (2004) and also attended a two-week Round Square Conference.
  • Intercollegiate - I represented IITB at several inter-collegiate speaking and literary competitions and won events like like Just-A-Minute (unbeaten last two years), courtroom trials, impromptus, quizzes, potpourris, wordgames, etc and won an IITB Instute Cultural Citation (given to about 5 of 1000+ graduating UGs/PGs) for the 30+ competitions I had won.
  • Interhostel - I represented my hostel in the Performing Arts Festival all four years (dance, background music, drama and voiceover, acting). I also took part in several inter-hostel competitions like dance, street play, drama, etc. I also represented my hostel in all literary and debating competitions, and got a hostel Cultural Citation for helping win the Literary Cup, and the overall interhostel Cultural Trophy.