Pierre-André Savalle  

I am currently working at Cisco. I was previously a Ph.D. candidate at Ecole Centrale Paris and ENS Cachan, with Gilles Fay and Nicolas Vayatis, in statistics and machine learning. I was previously at Penn where I worked on determinantal point processes. I also used to be interested in the synthesis of musical instruments using physical models.

Publications

• Untangling Local and Global Deformations in DCNNs for Image Classification and Sliding Window Detection
  • George Papandreou,
  • Iasonas Kokkinos,
  CVPR (oral)
  2015
• Interactions between Rank and Sparsity in Penalized Estimation, and Detection of Structured Objects
  Ph.D. thesis, Ecole Centrale Paris
  2014
• Localized Multiple Kernel Learning with Convex Hinge Loss Aggregation
  • Antoine Poliakov,
  preprint
  2014
• Detection of Correlations with Adaptive Sensing [poster]
  • Rui M. Castro,
  • Gabor Lugosi,
  IEEE Transactions on Information Theory
  2014
• Estimation of Simultaneously Sparse and Low Rank Matrices [poster]
  • Emile Richard,
  • Nicolas Vayatis,
  ICML
  2012
• Learning with Determinantal Point Processes
  M.Sc. thesis, University of Pennsylvania
  2011

Workshop papers

• Deformable Part Models with CNN Features [poster]
  • Stavros Tsogkas,
  • George Papandreou,
  • Iasonas Kokkinos,
  ECCV wksp. on Parts and Attributes
  2014
• Graph Denoising [poster]
  • Emile Richard,
  • Nicolas Vayatis,
  MMDS
  2012
• Graph Prediction in a Low-rank and Autoregressive Setting [poster]
  • Emile Richard,
  • Nicolas Vayatis,
  ICML wksp. on Sparsity
  2012