Last ten or so pubs of Manfred K. Warmuth

Last ten or so pubs of Manfred K. Warmuth



  • A case where a spindly two-layer linear network whips any neural network with a fully connected input layer arXiv

  • Rank-smoothed Pairwise Learning in Perceptual Quality Assessment. ICIP20 paper slides video

  • Reparameterizing Mirror Descent as Gradient Descent. NeurIIPS20 paper
    See also previous version which contains more material on matrix updates:
    Interpolating Between Gradient Descent and Exponentiated Gradient Using Reparameterized Gradient Descent arXiv,v1

  • Divergence-Based Motivation for Online EM and Combining Hidden Variable Models. UAI20 paper video slides

  • Winnowing with Gradient Descent. COLT20 paper video slides

  • TriMap: Large-scale dimensionality reduction using triplets arXiv
  • Also: A more globally accurate dimensionality reduction method using triplets arXiv

  • An Implicit Form of Krasulina's k-PCA Update without the Orthonormality Constraint. AAAI20 paper AAAI20 poster

  • Robust Bi-Tempered Logistic Loss Based on Bregman Divergences. Neurips19 paper poster talk

  • Unlabeled sample compression schemes and corner peelings for ample and maximum classes ICALP19 paper

  • Minimax experimental design: Bridging the gap between statistical and worst-case approaches to least squares regression COLT19 paper

  • Adaptive scale-invariant online algorithms for learning linear models ICML19 paper supplement

  • Two-temperature logistic regression based on the Tsallis divergence AISTAT19 paper supplement poster

  • Online Non-Additive Path Learning under Full and Partial Information, ALT19 paper talk

  • Unbiased estimators for random design regression. Expanded paper about the below two conference papers. journal submission

  • Correcting the bias in least squares regression with volume-rescaled sampling AISTAT19 paper supplement

  • Leveraged volume sampling for linear regression NeurIPS18 paper supplement

  • Reverse iterative volume sampling for linear regression
    Journal paper about the following 2 conference papers and more [JMLR paper]
    Long talk about volume sampling [talk]

  • Subsampling for Ridge Regression via Regularized Volume Sampling [AISTATS18 paper] [talk]

  • Unbiased estimates for linear regression via volume sampling [NeurIPS17 paper] [poster]

  • Online dynamic programming [NeurIPS17 paper]