Dr. Christoph Brauer

Contact Details

Address: Room: Phone: Email: Office Hours:

TU Braunschweig IAA Universitätsplatz 2 38106 Braunschweig 523 (Forum building) +49 (531) 391-7421 ch.brauer@tu-bs.de by arrangement

Publications

• Learning to dequantize speech signals by primal-dual networks: An approach for acoustic sensor networks
  With Ziyue Zhao, Dirk Lorenz and Tim Fingscheidt. 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 7000–7004. DOI: 10.1109/ICASSP.2019.8683341
• Complexity and Applications of the Homotopy Principle for Uniformly Constrained Sparse Minimization
  With Dirk A. Lorenz. Applied Mathematics and Optimization, 2019. DOI: 10.1007/s00245-019-09565-2 (full-text view-only version)
• A Primal-Dual Homotopy Algorithm for ℓ₁-Minimization with ℓ_∞-Constraints
  With Dirk A. Lorenz and Andreas M. Tillmann. Computational Optimization and Applications, 2018. DOI: 10.1007/s10589-018-9983-4 (Preprint: Optimization Online E-Print ID 2016-10-5700)
• Rank-Optimal Weighting or "How to be Best in the OECD Better Life Index?"
  With Dirk A. Lorenz and Jan Lorenz. Social Indicators Research, 2016, pp. 1–18. DOI: 10.1007/s11205-016-1416-0 (Preprint: arXiv:1608.04556)
• Sparse reconstruction of quantized speech signals
  With Dirk A. Lorenz and Timo Gerkmann. 2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 5940–5944. DOI: 10.1109/ICASSP.2016.7472817
• Cartoon-Texture-Noise Decomposition with Transport Norms
  With Dirk A. Lorenz. Scale Space and Variational Methods in Computer Vision: 5th International Conference, SSVM 2015, Lège-Cap Ferret, France, May 31 - June 4, 2015, Proceedings, pp. 142–153. DOI: 10.1007/978-3-319-18461-6_12

PhD thesis

• Homotopy Methods for Linear Optimization Problems with Sparsity Penalty and Applications
  Universitätsbibliothek Braunschweig, April 2018. DOI: 10.24355/dbbs.084-201804250929-0

Preprints

• A Sinkhorn-Newton method for entropic optimal transport
  With Christian Clason, Dirk Lorenz and Benedikt Wirth. arXiv:1710.06635
• Primal-dual residual networks
  With Dirk Lorenz. arXiv:1806.05823

GitHub

• ℓ₁-Houdini
  Matlab code related to the papers A Primal-Dual Homotopy Algorithm for ℓ₁-Minimization with ℓ_∞-Constraints and Complexity and Applications of the Homotopy Principle for Uniformly Constrained Sparse Minimization
• Primal-dual networks
  Python and TensorFlow code related to the paper Learning to dequantize speech signals by primal-dual networks: An approach for acoustic sensor networks

Conferences

• SPARS 2019, July 1-4, 2019, Toulouse, France
  Poster: Ergodic bilevel optimization (pdf, 224 KByte)
  Extended abstract: (pdf, 262 KByte)
  Poster: Exact recovery of partially sparse vectors (pdf, 996 KByte)
  Extended abstract: (pdf, 999 KByte)
• ICASSP 2019, May 12-17, 2019, Brighton, UK
  Poster: Learning to dequantize speech signals by primal-dual networks: An approach for acoustic sensor networks (pdf, 383 KByte)
• ISMP 2018, July 1-6, 2018, Bordeaux, France
  Talk: A primal-dual homotopy algorithm for sparse recovery with infinity norm constraints (pdf, 4 MByte)
• Workshop on Optimization, Machine Learning, and Data Science, April 12-13, 2018, Braunschweig, Germany
  Talk: A primal-dual homotopy algorithm for sparse recovery with infinity norm constraints (pdf, 4 MByte)
• NIPS 2017, December 4-9, 2017, Long Beach, USA
  Poster: A Sinkhorn-Newton method for entropic optimal transport (pdf, 211 KByte)
  Preprint: arXiv:1710.06635
• SPARS 2017, June 5-8, 2017, Lisbon, Portugal
  Poster: ℓ₁-Houdini: A New Homotopy Method for ℓ₁-Minimization (pdf, 308 KByte)
  Extended abstract: (pdf, 220 KByte)
• ICASSP 2016, March 20-25, 2016, Shanghai, China
  Poster: Sparse Reconstruction of Quantized Speech Signals (pdf, 218 KByte)
• GAMM 2016, March 7-11, 2016, Brunswick, Germany
  Talk: Sparse Reconstruction of Quantized Speech Signals (zip, 2 MByte)
• SPARS 2015, July 6-9, 2015, Cambridge, UK
  Poster: Heuristic Optimality Checks for Noise-Aware Sparse Recovery by ℓ₁-Minimization (pdf, 464 KByte)
  Extended abstract: (pdf, 487 KByte)
• SSVM 2015, May 31-June 4, 2015, Lège Cap Ferret, France
  Talk: Cartoon-Texture-Noise Decomposition with Transport Norms (pdf, 3 MByte)
• GAMM 2015, March 23-27, 2015, Lecce, Italy
  Talk: Cartoon-Texture-Noise Decomposition with Transport Norms (pdf, 3 MByte)
• OCIP 2015, March 9-11, 2015, Munich, Germany
  Talk: Cartoon-Texture-Noise Decomposition with Transport Norms (pdf, 2 MByte)

Teaching

• Upcoming: Winter 2019/20
  Maschinelles Lernen mit neuronalen Netzen
  Fortgeschrittenenpraktikum Data Science Poster (pdf, 7 MByte)
• Summer 2019
  Bachelor-Seminar Angewandte Mathematik
  Master-Seminar Angewandte Mathematik
• Winter 2018/19
  Maschinelles Lernen mit neuronalen Netzen Skript (pdf, 3 MByte)