Kontakt:

Telefon:	05121 / 883-40368
Telefax:	05121 / 883-40361
E-Mail:

Highlights:

• January 2016: I defended my PhD (Summa Cum Laude) with a dissertation titled "Invariant Features for Time-Series Classification", accessible here.

Teaching:

• Deep Learning, Summer Terms
• Business Analytics, Winter Terms

Reviewing and Program Committee:

• Journals
• Data Mining and Knowledge Discovery: 2014, 2015, 2017, 2018, 2019
• IEEE Transactions on Knowledge and Data Engineering (TKDE): 2018, 2019
• IEEE Transactions on Neural Networks and Learning Systems (TNNLS): 2019
• IEEE Access: 2018, 2019
• Knowledge-Based Systems: 2017
• Transportation Research (C): 2016
• Conferences
• European Conference on Machine Learning and Principles and Practice of Knowledge Discovery (ECML-PKDD): 2019
• International Joint Conference on Artificial Intelligence (IJCAI): 2019
• Association for the Advancement of Artificial Intelligence (AAAI): 2018
• IEEE International Conference on Intelligent Transportation Systems (ITSC): 2015, 2018

Publications:

Journal papers:

• Josif Grabocka, Nicolas Schilling, Lars Schmidt-Thieme (2016):
  Latent Time-Series Motifs, ACM Transactions on Knowledge Discovery from Data, TKDD
• Josif Grabocka, Martin Wistuba, Lars Schmidt-Thieme (2015):
  Fast Classification of Univariate and Multivariate Time series Through Shapelets Discovery, Journal of Knowledge and Information Systems, 5-year impact factor 2.02
• Josif Grabocka, Martin Wistuba, Lars Schmidt-Thieme (2014):
  Scalable Classification of Repetitive Time Series Through Frequencies of Local Polynomials, IEEE Transactions on Knowledge and Data Engineering, 5-year impact factor 2.87
• Josif Grabocka, Lars Schmidt-Thieme (2014):
  Invariant Time-Series Factorization, Journal of Data Mining and Knowledge Discovery, Impact Factor 2.77
• Josif Grabocka, Lars Schmidt-Thieme (2014):
  Learning Through Non-linearly Supervised Dimensionality Reduction, Springer Transactions on Large-Scale Data- and Knowledge-Centered Systems, LNCS

• Ekrem Oeztuerk, Fabio Ferreira, Hadi S. Jomaa, Lars Schmidt-Thieme, Josif Grabocka, Frank Hutter (2022):
  Zero-Shot AutoML with Pretrained Models, in Proceedings of the 39 th International Conference on Machine Learning.
• Shayan Jawed, Hadi S. Jomaa, Lars Schmidt-Thieme, Josif Grabocka (2021):
  Multi-task Learning Curve Forecasting Across Hyperparameter Configurations and Datasets, in European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD 2021).
• Ahmed Rashed, Josif Grabocka, Lars Schmidt-Thieme (2021):
  A Guided Learning Approach for Item Recommendation via Surrogate Loss Learning, in Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval.
• Hadi S. Jomaa, Josif Grabocka, Lars Schmidt-Thieme (2021):
  Dataset2Vec: Learning Dataset Meta-Features, in Data Mining and Knowledge Discovery Journal (The final publication is available at link.springer.com ). https://github.com/hadijomaa/dataset2vec .
• Jonas Sonntag, Lars Schmidt-Thieme, Josif Grabocka (2020):
  A machine learning approach to infer on-street parking occupancy based on parking meter transactions, in 2020 IEEE 23rd International Conference on Intelligent Transportation Systems (ITSC).
• Shayan Jawed, Josif Grabocka, Lars Schmidt-Thieme (2020):
  Self-Supervised Learning for Semi-Supervised Time Series Classification, in Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD 2020).
• Randolf Scholz, Josif Grabocka, Lars Schmidt-Thieme (2019):
  Learning surrogate losses, in arXiv preprint arXiv:1905.10108 (2019).
• Rafael Rêgo Drumond, Lukas Brinkmeyer, Josif Grabocka, Lars Schmidt-Thieme (2019):
  HIDRA: Head Initialization across Dynamic targets for Robust Architectures, in SIAM International Conference on Data Mining (SDM20), 2020..
• Lukas Brinkmeyer, Rafael Rêgo Drumond, Randolf Scholz, Josif Grabocka, Lars Schmidt-Thieme (2019):
  Chameleon: Learning Model Initializations Across Tasks With Different Schemas, in arXiv preprint arXiv:1909.13576 (2019).
• Hadi S. Jomaa, Josif Grabocka, Lars Schmidt-Thieme (2019):
  In Hindsight: A Smooth Reward for Steady Exploration, in arXiv preprint arXiv:1906.09781 (2019).
• Hadi S. Jomaa, Josif Grabocka, Lars Schmidt-Thieme (2019):
  Hyp-RL: Hyperparameter Optimization by Reinforcement Learning, in arXiv preprint arXiv:1906.11527 (2019).
• Vijaya Krishna Yalavarthi, Josif Grabocka, Hareesh Mandalapu, Lars Schmidt-Thieme (2019):
  Gait verification using deep learning with a pairwise loss, in 18th International Conference of the Biometrics Special Interest Group (BIOSIG 2019).
• Ahmed Rashed, Shayan Jawed, Jens Rehberg, Josif Grabocka, Lars Schmidt-Thieme, Andre Hintsches (2019):
  A Deep Multi-Task Approach for Residual Value Forecasting, in European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD 2019).
• Mohsan Jameel, Josif Grabocka, Mofassir ul Islam Arif, Lars Schmidt-Thieme (2019):
  Ring-Star : A Sparse Topology for Faster ModelAveraging in Decentralized Parallel SGD , in In Joint European Conference on Machine Learning and Knowledge Discovery in Databases (DMLE@ECML-PKDD 2019).
• Ahmed Rashed, Josif Grabocka, Lars Schmidt-Thieme (2019):
  Attribute-Aware Non-Linear Co-Embeddings of Graph Features, in 13th ACM Recommender Systems Conference (RecSys 2019).
• Ahmed Rashed, Josif Grabocka, Lars Schmidt-Thieme (2019):
  Multi-Relational Classification via Bayesian Ranked Non-Linear Embeddings, in 25th ACM SIGKDD conference on knowledge discovery and data mining (SIGKDD 2019). Acceptance Rate: 14.1% (170 out of 1200).
• Hadi S. Jomaa, Josif Grabocka, Lars Schmidt-Thieme (2019):
  A Hybrid Convolutional Approach for Parking Availability Prediction, in International Conference on Convolutional Neural Networks (IJCNN2019) .
• Ahmed Rashed, Josif Grabocka, Lars Schmidt-Thieme (2019):
  Multi-Label Network Classification via Weighted Personalized Factorizations, in International Conference on Agents and Artificial Intelligence (ICAART 2019).
• Josif Grabocka, Lars Schmidt-Thieme (2018):
  Neuralwarp: Time-series similarity with warping networks, in arXiv preprint arXiv:1812.08306.
• Shayan Jawed, Eya Boumaiza, Josif Grabocka, Lars Schmidt-Thieme (2018):
  Data-Driven Vehicle Trajectory Forecasting, in European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases Workshop on (KNOWMe@ECML PKDD 2018).
• Dripta S Raychaudhuri, Josif Grabocka, Lars Schmidt-Thieme (2017):
  Channel masking for multivariate time series shapelets, in arXiv preprint arXiv:1711.00812 (2017).
• Hanh T. H. Nguyen, Martin Wistuba, Josif Grabocka, Lucas Rego Drumond, Lars Schmidt-Thieme (2017):
  Personalized Deep Learning for Tag Recommendation, in Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD 2017), Jeju, South Korea.
• Mit Shah, Josif Grabocka, Nicolas Schilling, Martin Wistuba, Lars Schmidt-Thieme (2016):
  Learning DTW-Shapelets for Time-Series Classification, in ACM IKDD Conference on Data Science. Best Paper Award.
• Josif Grabocka, Alexandros Dalkalitsis, Athanasios Lois, Evangelos Katsaros, Lars Schmidt-Thieme (2014):
  Realistic Optimal Policies for Energy-Efficient Train Driving, in Proceedings of the 17th International IEEE Conference on Intelligent Transportation Systems, ITSC 2014.
• Josif Grabocka, Nicolas Schilling, Martin Wistuba, Lars Schmidt-Thieme (2014):
  Learning Time-Series Shapelets, in Proceedings of the 20th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, KDD 2014. Acceptance Rate: 14.6% (151 out of 1036).
• Josif Grabocka, Erind Bedalli, Lars Schmidt-Thieme (2014):
  Supervised Nonlinear Factorizations Excel In Semi-supervised Regression , in Proceedings of the 18th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2014, Springer, LNCS, Tainan, Taiwan.
• Josif Grabocka, Lucas Drumond, Lars Schmidt-Thieme (2013):
  Supervised Dimensionality Reduction Via Nonlinear Target Estimation, in Proceedings of the 15th International Conference on Data Warehousing and Knowledge Discovery, DaWaK 2013 .
• Josif Grabocka, Erind Bedalli, Lars Schmidt-Thieme (2012):
  Efficient Classification of Long Time Series, in Proceedings of ICT Innovations Conference 2012, Advances in Intelligent Systems and Computing, Volume 207, pp 47-57, Springer, Berlin/Heidelberg .
• Josif Grabocka, Alexandros Nanopoulos, Lars Schmidt-Thieme (2012):
  Invariant Time-Series Classification, in Proceedings of European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML'12) , Bristol, United Kingdom.
• Josif Grabocka, Alexandros Nanopoulos, Lars Schmidt-Thieme (2012):
  Classification of Sparse Time Series via Supervised Matrix Factorization, in Proceedings of the 26th AAAI Conference on Artificial Intelligence (AAAI'12), Toronto, Canada.