pith. sign in

Matthias Hein

Identifiers

  • name variant Matthias Hein 0.60 · backfill

Papers (47)

  1. Preference Redirection via Attention Concentration: An Attack on Computer Use Agents cs.LG · 2026 · author #2
  2. Perturb and Recover: Fine-tuning for Effective Backdoor Removal from CLIP cs.LG · 2024 · author #3
  3. Spectral Clustering of Signed Graphs via Matrix Power Means cs.LG · 2019 · author #3
  4. Why ReLU networks yield high-confidence predictions far away from the training data and how to mitigate the problem cs.LG · 2018 · author #1
  5. Disentangling Adversarial Robustness and Generalization cs.CV · 2018 · author #2
  6. A randomized gradient-free attack on ReLU networks cs.LG · 2018 · author #2
  7. Logit Pairing Methods Can Fool Gradient-Based Attacks cs.LG · 2018 · author #4
  8. On the loss landscape of a class of deep neural networks with no bad local valleys cs.LG · 2018 · author #3
  9. Neural Networks Should Be Wide Enough to Learn Disconnected Decision Regions cs.LG · 2018 · author #3
  10. Error estimates for spectral convergence of the graph Laplacian on random geometric graphs towards the Laplace--Beltrami operator stat.ML · 2018 · author #3
  11. Optimization Landscape and Expressivity of Deep CNNs cs.LG · 2017 · author #2
  12. Community detection in networks via nonlinear modularity eigenvectors cs.SI · 2017 · author #3
  13. Variants of RMSProp and Adagrad with Logarithmic Regret Bounds cs.LG · 2017 · author #2
  14. Formal Guarantees on the Robustness of a Classifier against Adversarial Manipulation cs.LG · 2017 · author #1
  15. The loss surface of deep and wide neural networks cs.LG · 2017 · author #2
  16. The Perron-Frobenius Theorem for Multi-homogeneous Maps math.SP · 2017 · author #3
  17. Clustering Signed Networks with the Geometric Mean of Laplacians stat.ML · 2017 · author #3
  18. Analysis and Optimization of Loss Functions for Multiclass, Top-k, and Multilabel Classification cs.CV · 2016 · author #2
  19. Globally Optimal Training of Generalized Polynomial Neural Networks with Nonlinear Spectral Methods cs.LG · 2016 · author #3
  20. Improved Image Boundaries for Better Video Segmentation cs.CV · 2016 · author #4
  21. Latent Embeddings for Zero-shot Classification cs.CV · 2016 · author #5
  22. Simple Does It: Weakly Supervised Instance and Semantic Segmentation cs.CV · 2016 · author #4
  23. A nodal domain theorem and a higher-order Cheeger inequality for the graph $p$-Laplacian math.SP · 2016 · author #2
  24. Loss Functions for Top-k Error: Analysis and Insights stat.ML · 2015 · author #2
  25. Weakly Supervised Object Boundaries cs.CV · 2015 · author #4
  26. Top-k Multiclass SVM stat.ML · 2015 · author #2
  27. Efficient Output Kernel Learning for Multiple Tasks stat.ML · 2015 · author #3
  28. An Efficient Multilinear Optimization Framework for Hypergraph Matching cs.CV · 2015 · author #4
  29. Robust PCA: Optimization of the Robust Reconstruction Error over the Stiefel Manifold stat.ML · 2015 · author #3
  30. Towards Realistic Team Formation in Social Networks based on Densest Subgraphs cs.SI · 2015 · author #3
  31. Constrained 1-Spectral Clustering stat.ML · 2015 · author #2
  32. Tight Continuous Relaxation of the Balanced $k$-Cut Problem stat.ML · 2015 · author #3
  33. Regularization-free estimation in trace regression with symmetric positive semidefinite matrices stat.ML · 2015 · author #3
  34. Tensor norm and maximal singular vectors of non-negative tensors - a Perron-Frobenius theorem, a Collatz-Wielandt characterization and a generalized power method math.SP · 2015 · author #2
  35. Estimation of positive definite M-matrices and structure learning for attractive Gaussian Markov Random fields math.ST · 2014 · author #2
  36. Matrix factorization with Binary Components stat.ML · 2014 · author #2
  37. Nonlinear Eigenproblems in Data Analysis - Balanced Graph Cuts and the RatioDCA-Prox stat.ML · 2013 · author #3
  38. The Total Variation on Hypergraphs - Learning on Hypergraphs Revisited stat.ML · 2013 · author #1
  39. Constrained fractional set programs and their application in local clustering and community detection stat.ML · 2013 · author #4
  40. Learning Using Privileged Information: SVM+ and Weighted SVM stat.ML · 2013 · author #2
  41. Non-negative least squares for high-dimensional linear models: consistency and sparse recovery without regularization math.ST · 2012 · author #2
  42. How the result of graph clustering methods depends on the construction of the graph stat.ML · 2011 · author #3
  43. An Inverse Power Method for Nonlinear Eigenproblems with Applications in 1-Spectral Clustering and Sparse PCA cs.LG · 2010 · author #1
  44. Hitting and commute times in large graphs are often misleading cs.DS · 2010 · author #3
  45. Optimal construction of k-nearest neighbor graphs for identifying noisy clusters stat.ML · 2009 · author #2
  46. Graph Laplacians and their convergence on random neighborhood graphs math.ST · 2006 · author #1
  47. Numerical evolution of axisymmetric, isolated systems in General Relativity gr-qc · 2002 · author #2

Mentions

  • 1702.03230 #3 · arxiv_oai · confidence 0.70 Matthias Hein
  • 1701.00757 #3 · arxiv_oai · confidence 0.70 Matthias Hein
  • 1503.01273 #2 · arxiv_oai · confidence 0.70 Matthias Hein
  • 1506.00323 #3 · backfill · confidence 0.70 Matthias Hein
  • 1505.06661 #3 · backfill · confidence 0.70 Matthias Hein
  • 1505.06485 #2 · backfill · confidence 0.70 Matthias Hein
  • 1505.06478 #3 · backfill · confidence 0.70 Matthias Hein
  • 1504.06305 #3 · backfill · confidence 0.70 Matthias Hein
  • 1503.01273 #2 · backfill · confidence 0.70 Matthias Hein
  • 1404.6640 #2 · backfill · confidence 0.70 Matthias Hein
  • 1401.6024 #2 · backfill · confidence 0.70 Matthias Hein
  • 1312.5192 #3 · backfill · confidence 0.70 Matthias Hein
  • 1312.5179 #1 · backfill · confidence 0.70 Matthias Hein
  • 1306.3409 #4 · backfill · confidence 0.70 Matthias Hein
  • 1306.3161 #2 · backfill · confidence 0.70 Matthias Hein
  • 1205.0953 #2 · backfill · confidence 0.70 Matthias Hein
  • 1102.2075 #3 · backfill · confidence 0.70 Matthias Hein
  • 1012.0774 #1 · backfill · confidence 0.70 Matthias Hein
  • 1003.1266 #3 · backfill · confidence 0.70 Matthias Hein
  • 0912.3408 #2 · backfill · confidence 0.70 Matthias Hein

Frequent Coauthors