• (31.01.2020) Winner announcement: We have 3 winner of the 123ai.de mechanical mascot! Congrats


  • (13.12.2019) First giveaway announcement!
  • (11.12.2019) First weekly newsletter!
  • (02.12.2019-06.12.2019) I organized the corporate training week on Bayesian Learning for Petrobras in Rio de Janeiro
  • (09.09.2019) 123AI.de funding starts
  • (31.08.2019) Last day as a post doc at the machine learning group at the TU Berlin
  • (24.04.2019-08.05.2019) I organized a 2-week workshop “Introduction to Machine Learning” at Petrobras in Rio de Janeiro and delivered the lectures on Bayesian learning and variational inference
  • (01.03.2019) Paper on “Deep Semi-Supervised Anomaly Detection” is now available in arXiv.


  • (22.12.2018) Our paper on “Partial Optimality of Dual Decomposition for MAP Inference in Pairwise MRFs” has been accepted at AISTATS 2019
  • (26.11.2018-30.11.2018) I gave a 4-day workshop on “Introduction to Anomaly Detection” at Petrobras in Rio de Janeiro
  • (29.06.2018) We have a new arXiv paper out “Unsupervised Detection and Explanation of Latent-class Contextual Anomalies
  • (07.03.2018) Our ICML paper on “Deep one-class classification” is out in the wild


  • (15.12.2017) I jotted down some thoughts about explanation and interpretation of machine learning models. The main idea was to categorize approaches into a meaningful taxonomy. The blog post can be found here
  • (01.10.2017) Our paper “Ensembles of Lasso Screening Rules” (Seunghak Lee, Nico Görnitz, Eric P. Xing, David Heckerman, and Christoph Lippert) is finally accepted for publication in TPAMI!
  • (01.08.2017) Our paper “Support vector data descriptions and k-means clustering: one class?,” is accepted for publication in IEEE TNNLS
  • (02.06.2017) Our paper “Minimizing Trust Leaks for Robust Sybil Detection” is accepted for publication at ICML
  • (20.05.2017) Our paper Transductive Regression for Data with Latent Dependency Structure” is accepted for publication in IEEE TNNLS
  • (19.05.2017) …and parental leave starts!
  • (03.05.2017) Our paper “Porosity Estimation by Semi-supervised Learning with Sparsely Available Labeled Samples” is accepted for publication in Computers and Geosciences
  • (27.03.2017) Our paper “ML2Motif—Reliable extraction of discriminative sequence motifs from learning machines” is accepted for publication in PLoS one!


  • (24.11.2016) Our paper “Feature Importance Measure for Non-linear Learning Algorithms” won the best paper award at the Interpretable ML for Complex Systems (NIPS Workshop)!
  • (19.06.2016) We setup an online tool for collecting your questions to the panel for the ICML anomaly detection workshop. You can also up-vote questions. The list can be accessed here
  • (06.06.2016) Registration for ICML is back online (Tickets for tutorials, conference, and workshops still available)
  • (03.06.2016) Only 3 weeks to go for our workshop (unfortunately workshop tickets for the ICML 2016 are sold-out now)
  • (09.04.2016) added the CfP for our ICML 2016 Workshop on Anomaly Detection
  • (03.01.2016) here is how you contact me and some additional basic information about me and my research and stuff
  • (03.01.2016) just started the site, please bear with me…