Aaditya Ramdas – Publications


Anytime-valid, safe confidence intervals and p-values (package) (tutorial)

  • Admissible anytime-valid sequential inference must rely on nonnegative martingales
    A. Ramdas, J. Ruf, M. Larsson, W. Koolen       arxiv  

  • Testing exchangeability: fork-convexity, supermartingales, and e-processes
    A. Ramdas, J. Ruf, M. Larsson, W. Koolen       Intl J of Approximate Reasoning, 2021   arxiv   proc

  • Estimating means of bounded random variables by betting
    I. Waudby-Smith, A. Ramdas       (JRSSB, submitted)   arxiv  

  • Comparing sequential forecasters
    Y.J. Choe, A. Ramdas       arxiv  

  • RiLACS: Risk-limiting audits via confidence sequences
    I. Waudby-Smith, P. Stark, A. Ramdas       EVoteID, 2021   arxiv   (Best Paper award: Security, Usability and Technical Track)

  • Sequential estimation of convex divergences using reverse submartingales and exchangeable filtrations
    T. Manole, A. Ramdas       arxiv   video

  • Off-policy confidence sequences
    N. Karampatziakis, P. Mineiro, A. Ramdas       ICML, 2021   arxiv  

  • Doubly robust confidence sequences for sequential causal inference
    I. Waudby-Smith, D. Arbour, R. Sinha, E. Kennedy, A. Ramdas       arxiv   package  

  • Confidence sequences for sampling without replacement
    I. Waudby-Smith, A. Ramdas       NeurIPS, 2020   arxiv   proc

  • Time-uniform, nonparametric, nonasymptotic confidence sequences
    S. Howard, A. Ramdas, J. Sekhon, J. McAuliffe       Annals of Stat., 2021   arxiv   proc   code   tutorial

  • Time-uniform Chernoff bounds via nonnegative supermartingales
    S. Howard, A. Ramdas, J. Sekhon, J. McAuliffe       Prob. Surveys, 2020   arxiv   proc   talk

  • Universal inference
    L. Wasserman, A. Ramdas, S. Balakrishnan       PNAS, 2020   arxiv   proc   talk

  • Sequential nonparametric testing with the law of the iterated logarithm
    A. Balsubramani*, A. Ramdas*       Uncertainty in AI (UAI), 2016   arxiv   proc   errata

  • Nonparametric iterated-logarithm extensions of the sequential generalized likelihood ratio test
    J. Shin, A. Ramdas, A. Rinaldo       IEEE J. on Selected Areas in Info. Theory, 2021   arxiv

  • Uncertainty quantification using martingales for misspecified Gaussian processes
    W. Neiswanger, A. Ramdas       Algorithmic Learning Theory (ALT), 2021   arxiv   code   talk   proc

  • Sequential estimation of quantiles with applications to A/B-testing and best-arm identification
    S. Howard, A. Ramdas       Bernoulli, 2022   arxiv   code   talk


Multi-armed bandits

  • A unified framework for bandit multiple testing
    Z. Xu, R. Wang, A. Ramdas       NeurIPS, 2021   arxiv  

  • On conditional versus marginal bias in multi-armed bandits
    J. Shin, A. Ramdas, A. Rinaldo       ICML, 2020   arxiv

  • Are sample means in multi-armed bandits positively or negatively biased?
    J. Shin, A. Ramdas, A. Rinaldo       NeurIPS, 2019   arxiv   poster   proc

  • On the bias, risk and consistency of sample means in multi-armed bandits
    J. Shin, A. Ramdas, A. Rinaldo       SIAM J Math of Data Science (SIMODS), 2022   arxiv   talk

  • MAB-FDR: Multi (A)rmed/(B)andit testing with online FDR control
    F. Yang, A. Ramdas, K. Jamieson, M. Wainwright       NeurIPS, 2017   arxiv   code   30-min talk   proc   (spotlight talk)


Online multiple testing (package) (vignette) (shiny app, FDR) (shiny app, FWER)

  • ADDIS: an adaptive discarding algorithm for online FDR control with conservative nulls
    J. Tian, A. Ramdas       NeurIPS, 2019   arxiv   code

  • Online control of the false coverage rate and false sign rate
    A. Weinstein*, A. Ramdas*       ICML, 2020   arxiv   proc

  • Online control of the familywise error rate
    J. Tian*, A. Ramdas*       Statistical Methods in Medical Research, 2021   arxiv   proc

  • Asynchronous online testing of multiple hypotheses
    T. Zrnic, A. Ramdas, M. Jordan       JMLR, 2021   arxiv   code   blog   proc

  • The power of batching in multiple hypothesis testing
    T. Zrnic, D. Jiang, A. Ramdas, M. Jordan       AISTATS, 2020   arxiv   talk   proc

  • SAFFRON: an adaptive algorithm for online FDR control
    A. Ramdas, T. Zrnic, M. Wainwright, M. Jordan       ICML, 2018   arxiv   proc   github   (full oral talk)

  • Online control of the false discovery rate with decaying memory
    A. Ramdas, F. Yang, M. Wainwright, M. Jordan       NeurIPS, 2017   arxiv   summary   poster   proc   15-min from 44:00   (full oral talk)

  • Dynamic Algorithms for Online Multiple Testing
    Z. Xu, A. Ramdas       Mathematical and Scientific ML, 2021   arxiv


Interactive/dynamic multiple testing

  • Interactive identification of individuals with positive treatment effect while controlling false discoveries
    B. Duan, L. Wasserman, A. Ramdas       arxiv  

  • STAR: A general interactive framework for FDR control under structural constraints
    L. Lei, A. Ramdas, W. Fithian       Biometrika, 2020   arxiv   proc   poster   code

  • Familywise error rate control by interactive unmasking
    B. Duan, A. Ramdas, L. Wasserman       ICML, 2020   arxiv   code

  • Interactive martingale tests for the global null
    B. Duan, A. Ramdas, S. Balakrishnan, L. Wasserman       Electronic J. of Stat., 2020   arxiv   code   proc

  • Which Wilcoxon should we use? An interactive rank test and other alternatives
    B. Duan, A. Ramdas, L. Wasserman       arxiv

  • Simultaneous high-probability bounds on the FDP in structured, regression and online settings
    E. Katsevich, A. Ramdas       Annals of Stat., 2020   arxiv   code   proc

  • Large-scale simultaneous inference under dependence
    J. Tian, X. Chen, E. Katsevich, J. Goeman, A. Ramdas       arxiv  

  • QuTE: decentralized FDR control on sensor networks
    A. Ramdas, J. Chen, M. Wainwright, M. Jordan       IEEE CDC, 2017   code   proc   poster


Offline multiple testing

  • False discovery rate control with e-values
    R. Wang, A. Ramdas       (JRSSB, minor revision)   arxiv

  • Fast and powerful conditional randomization testing via distillation
    M. Liu, E. Katsevich, L. Janson, A. Ramdas       Biometrika, 2021   arxiv   code

  • A unified treatment of multiple testing with prior knowledge using the p-filter
    A. Ramdas, R. F. Barber, M. Wainwright, M. Jordan       Annals of Stat., 2019   arxiv   code   proc

  • DAGGER: A sequential algorithm for FDR control on DAGs
    A. Ramdas, J. Chen, M. Wainwright, M. Jordan       Biometrika, 2019   arxiv   code   proc

  • Optimal rates and tradeoffs in multiple testing
    M. Rabinovich, A. Ramdas, M. Wainwright, M. Jordan       Stat. Sinica, 2019   arxiv   poster   proc

  • p-filter: multi-layer FDR control for grouped hypotheses
    R. F. Barber*, A. Ramdas*       JRSSB, 2016   arxiv   code   proc   poster


High-dimensional/nonparametric testing

  • Dimension-agnostic inference
    I. Kim, A. Ramdas       (Annals of Stat., submitted)   arxiv

  • Gaussian Universal Likelihood Ratio Testing
    R. Dunn, A. Ramdas, S. Balakrishnan, L. Wasserman       arxiv  

  • On the power of conditional independence testing under model-X
    E. Katsevich, A. Ramdas       arxiv

  • Adaptivity & computation-statistics tradeoffs for kernel & distance based high-dimensional two sample testing
    A. Ramdas, S. Reddi, B. Poczos, A. Singh, L. Wasserman       arxiv   poster  

  • Classification accuracy as a proxy for two sample testing
    I. Kim*, A. Ramdas*, A. Singh, L. Wasserman       Annals of Stat., 2021   arxiv   proc

  • Minimax lower bounds for linear independence testing
    D. Isenberg*, A. Ramdas*, A. Singh, L. Wasserman       IEEE ISIT, 2016   arxiv   proc

  • 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, 2015   proc   arxiv   errata

  • 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, 2015   proc   arxiv   supp

  • A higher order Kolmogorov Smirnov test
    V. Sadhanala, Y. Wang, A. Ramdas, R. Tibshirani       AISTATS, 2019   arxiv   proc

  • On Wasserstein two sample testing and related families of nonparametric tests
    A. Ramdas*, N. Garcia*, M. Cuturi       Entropy, 2017 (Special Issue)   arxiv   proc   SI

  • Generative models and model criticism via optimized Maximum Mean Discrepancy
    D. Sutherland, H. Tung, H. Strathmann, S. De, A. Ramdas, A. Smola, A. Gretton       ICLR, 2017   arxiv   proc   poster   code

  • Fast two-sample testing with analytic representations of probability measures
    K. Chwialkowski, A. Ramdas, D. Sejdinovic, A. Gretton       NeurIPS, 2015   arxiv   github   proc

  • Nonparametric independence testing for small sample sizes
    A. Ramdas*, L. Wehbe*       IJCAI, 2015   arxiv   proc   (oral talk)


Distribution-free uncertainty quantification (conformal, calibration) (package) (package 2) (tutorial)

  • Top-label calibration
    C. Gupta, A. Ramdas       arxiv  

  • Distribution-free calibration guarantees for histogram binning without sample splitting
    C. Gupta, A. Ramdas       ICML, 2021   arxiv   proc

  • Distribution-free uncertainty quantification for classification under label shift
    A. Podkopaev, A. Ramdas       UAI, 2021   arxiv  

  • Distribution-free binary classification: prediction sets, confidence intervals and calibration
    C. Gupta, A. Podkopaev, A. Ramdas       NeurIPS, 2020   arxiv   proc   talk

  • Nested conformal prediction and quantile out-of-bag ensemble methods
    C. Gupta, A. Kuchibhotla, A. Ramdas       Pattern Recognition, Special Issue on Conformal Prediction   arxiv   code   talk

  • The limits of distribution-free conditional predictive inference
    R. Barber, E. Candes, A. Ramdas, R. Tibshirani       Information and Inference, 2020   arxiv   proc

  • Conformal prediction under covariate shift
    R. Tibshirani, R. Barber, E. Candes, A. Ramdas       NeurIPS, 2019   arxiv   proc   poster  

  • Predictive inference with the jackknife+
    R. Barber, E. Candes, A. Ramdas, R. Tibshirani       Annals of Stat., 2021   arxiv   code   proc


Convex optimization

  • Path length bounds for gradient descent and flow
    C. Gupta, S. Balakrishnan, A. Ramdas       JMLR, 2021   arxiv   proc   blog

  • Iterative methods for solving factorized linear systems
    A. Ma, D. Needell, A. Ramdas       SIAM J on Matrix Anal. and App., 2017   arxiv   proc

  • Rows vs columns : randomized Kaczmarz or Gauss-Seidel for ridge regression
    A. Hefny*, D. Needell*, A. Ramdas*       SIAM J on Sci. Comp., 2017   arxiv   proc

  • Convergence properties of the randomized extended Gauss-Seidel and Kaczmarz methods
    A. Ma*, D. Needell*, A. Ramdas*       SIAM J on Matrix Anal. and App., 2015   arxiv   proc   code

  • Towards a deeper geometric, analytic and algorithmic understanding of margins
    A. Ramdas, J. Pena       Opt. Methods and Software, 2015   arxiv   proc

  • Margins, kernels and non-linear smoothed perceptrons
    A. Ramdas, J. Pena       ICML, 2014   arxiv   proc   poster   (oral talk)

  • Algorithmic connections between active learning and stochastic convex optimization
    A. Ramdas, A. Singh       Alg. Learning Theory, 2013   arxiv   proc   poster  

  • Optimal rates for stochastic convex optimization under Tsybakov's noise condition
    A. Ramdas, A. Singh       ICML, 2013   proc   arxiv   poster   (oral talk)


Scientific collaborations

  • Brainprint: identifying individuals from Magnetoencephalography
    S. Wu, A. Ramdas, L. Wehbe       (submitted, Nature Communications)   bioRxiv

  • Regularized brain reading with shrinkage and smoothing
    L. Wehbe, A. Ramdas, R. Steorts, C. Shalizi       Annals of App. Stat., 2015   arxiv   proc

  • 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       PLoS ONE, 2014   website   proc  


One-off projects

  • Analyzing student strategies in blended courses using clickstream data
    N. Akpinar, A. Ramdas, U. Acar       Educational Data Mining, 2020   arxiv   talk  

  • Function-specific mixing times and concentration away from equilibrium
    M. Rabinovich, A. Ramdas, M. Wainwright, M. Jordan       Bayesian Analysis, 2019   arxiv   proc   poster  

  • Decoding from pooled data (II): sharp information-theoretic bounds
    A. El-Alaoui, A. Ramdas, F. Krzakala, L. Zdeborova, M. Jordan       SIAM J on Math. of Data Science, 2019   arxiv   proc

  • Decoding from pooled data (I): phase transitions of message passing
    A. El-Alaoui, A. Ramdas, F. Krzakala, L. Zdeborova, M. Jordan       IEEE Trans. on Info. Theory, 2018   arxiv   proc

  • On the power of online thinning in reducing discrepancy
    R. Dwivedi, O. N. Feldheim, Ori Gurel-Gurevich, A. Ramdas       Prob. Theory and Related Fields, 2018   arxiv   poster   proc

  • On kernel methods for covariates that are rankings
    H. Mania, A. Ramdas, M. Wainwright, M. Jordan, B. Recht       Electronic J of Stat., 2018   arxiv   proc

  • Asymptotic behavior of Lq-based Laplacian regularization in semi-supervised learning
    A. El-Alaoui, X. Cheng, A. Ramdas, M. Wainwright, M. Jordan       COLT, 2016   arxiv   proc

  • An analysis of active learning with uniform feature noise
    A. Ramdas, A. Singh, L. Wasserman, B. Poczos       AISTATS, 2014   arxiv   proc   poster   (oral talk)  


Miscellaneous

  • Discussion: Covariate-assisted ranking and screening for large-scale two-sample inference
    A. Ramdas       JRSSB, 2019   proc

  • Algorithms for graph similarity and subgraph matching
    D. Koutra, A. Parikh, A. Ramdas, J. Xiang       Tech Report   pdf