Bastian Alt
PhD student
Contact
bastian.alt@bcs.tu-...
work +49 6151 16-57 246
fax +49 6151 16-57 241
Work
S3/19 1.8
Rundeturmstraße 12
64283
Darmstadt
Bastian Alt received his B.Sc. and M.Sc. degrees in Electrical Engineering and Information Technology from Technische Universität Darmstadt in November 2013 and December 2016, respectively. His master thesis was entitled “Robust and Sparse Regression for Linear Inverse Problems”. He joined the Bioinspired Communication Systems Lab in January 2017.
Bastian is part of the Collaborative Research Centre (CRC) 1053 “MAKI– Multi-Mechanisms Adaptation for the Future Internet” in sub-project B4. He is interested in modelling of network systems. For this tools from machine learning/ statistics, control and optimization are used.
His research interests include topics from machine learning and control such as Bayesian inference of Markov chains, Bayesian non-parametrics, Bayesian optimization, stochastic optimal control, planning in Markov Decision Processes (MDPs), reinforcement learning and bandit algorithms.
Possible applications include learning and control of communication systems, e.g., in queuing networks.
Possible Student Projects you can find here.
2020
Kar, Sounak and Rehrmann, Robin and Mukhopadhyay, Arpan and Alt, Bastian and Ciucu, Florin and Koeppl, Heinz and Carsten, Binnig and Rizk, Amr (2020):
On the Throughput Optimization in Large-Scale Batch-Processing Systems.
pp. 1:15, 38th International Symposium on Computer Performance, Modeling, Measurements and Evaluation (IFIP 2020), virtual Conference, 02.-06.11., [Conference or Workshop Item]
Alt, B. and Schultheis, M. and Koeppl, H. (2020):
POMDPs in Continuous Time and Discrete Spaces.
34th Conference on Neural Information Processing Systems (NeurIPS 2020), virtual Conference, 06.-12.12., [Conference or Workshop Item]
Machkour, J. and Muma, M. and Alt, B. and Zoubir, A. M. (2020):
A Robust Adaptive Lasso Estimator for the Independent Contamination Model.
In: Signal Processing, 174, Elsevier ScienceDirect, ISSN 01651684,
DOI: 10.1016/j.sigpro.2020.107608,
[Article]
KhudaBukhsh, W. R. and Kar, S. and Alt, B. and Rizk, A. and Koeppl, H. (2020):
Generalized Cost-Based Job Scheduling in Very Large Heterogenous Cluster Systems.
In: IEEE Transactions on Parallel and Distributed Systems (TPDS), 31 (11), pp. 2594-2604. IEEE, ISSN 1045-9219, e-ISSN 1045-9219,
DOI: 10.1109/TPDS.2020.2997771,
[Article]
2019
Alt, B. and Šošić, A. and Koeppl, H. (2019):
Correlation Priors for Reinforcement Learning.
33rd Conference on Neural Information Processing Systems (NeurIPS 2019), Vancouver, Kanada, 09.12.-13.12.2019, [Conference or Workshop Item]
Alt, Bastian and Ballard, Trevor and Steinmetz, Ralf and Koeppl, Heinz and Rizk, Amr (2019):
CBA: Contextual Quality Adaptation for Adaptive Bitrate Video Streaming.
pp. 1000-1008, IEEE, IEEE Conference on Computer Communications (INFOCOM 2019), Paris, France, 29.04.-02.05., ISBN 978-1-7281-0516-1,
DOI: 10.1109/INFOCOM.2019.8737418,
[Conference or Workshop Item]
Alt, Bastian and Weckesser, Markus and Becker, Christian and Hollick, Matthias and Kar, Sounak and Klein, Anja and Klose, Robin and Kluge, Roland and Koeppl, Heinz and KhudaBukhsh, Wasiur R. and Luthra, Manisha and Mousavi, Mahdi and Mühlhäuser, Max and Pfannemüller, Martin and Rizk, Amr and Schürr, Andy and Steinmetz, Ralf (2019):
Transitions: A Protocol-Independent View of the Future Internet.
In: Proceedings of the IEEE, 107 (4), pp. 835-846. ISSN 0018-9219,
DOI: 10.1109/JPROC.2019.2895964,
[Article]
2018
Alt, Bastian and Messer, Michael and Roeper, Jochen and Schneider, Gaby and Koeppl, Heinz (2018):
Non-Parametric Bayesian Inference for Change Point Detection in Neural Spike Trains.
2018 IEEE Statistical Signal Processing Workshop (SSP 2018), Freiburg im Breisgau, Germany, 10.-13.06., [Conference or Workshop Item]
KhudaBukhsh, Wasiur R. and Alt, Bastian and Kar, Sounak and Rizk, Amr and Koeppl, Heinz (2018):
Collaborative Uploading in Heterogeneous Networks: Optimal and Adaptive Strategies.
INFOCOM 2018 - IEEE Conference on Computer Communications, Honolulu, USA, 16.-19.04., DOI: 10.1109/INFOCOM.2018.8486310,
[Conference or Workshop Item]
2017
Machkour, J. and Alt, B. and Muma, M. and Zoubir, A. M. (2017):
The Outlier-Corrected-Data-Adaptive Lasso: A New Robust Estimator for the Independent Contamination Model.
pp. 1699-1703, European Signal Processing Conference 2017 (EUSIPCO 2017), Kos, Greece, 28.08.-02.09., DOI: 10.23919/EUSIPCO.2017.8081489,
[Conference or Workshop Item]
Current Student Projects
Completed Student Projects
|
Gabriel Krack (Bachelor thesis) |
Stochastic Model Predictive Control for Quadcopters | 10/2020 |
|
Erik Samstag (Bachelor thesis – co-supervision w/ IAS TU Darmstadt) |
Bayesian Optimization for Biochemical Dose Response Measurements | 10/2020 |
|
Michael Suffel (Master thesis) |
A Decentralized Multi-Agent Model Predictive Control Framework for Quadcopters | 06/2020 |
|
Florian Swienty (Master thesis) |
Decentralized Collision-Free Flight of a Quadcopter Swarm | 04/2020 |
|
Lu Wan (Master thesis) |
Path Integral Control for Quadcopters | 03/2020 |
|
Sebastian Wirth (Bachelor thesis) |
State Space Reduction Using Variational Autoencoders | 12/2019 |
|
Intisar Naheen (Master thesis) |
Bayesian Reinforcement Learning with Maximum Entropy Finite State Controllers | 12/2018 |
|
Hans Stenglein (Bachelor thesis) |
Optimal Planning with Model Uncertainty | 11/2018 |
|
Hameer Abbasi (Master thesis – w/ Wasiur KhudaBukhsh) |
Decentralised Decision Making for Queueing Systems | 08/2018 |
Co-Supervised Student Projects
|
Sakshi Goyal (Master thesis – co-supervison w/ Visual Inference TU Darmstadt) |
Instance Segmentation with Group Proposals | |
|
Yue Wang (Master thesis – co-supervison w/ Ubiquitous Knowledge Processing (UKP) Lab TU Darmstadt) |
Addressing The Speakers’ Persona Consistency Issue in Neural Conversational Agents | 12/2019 |
|
Sumanth Venugopal (Master thesis – co-supervison w/ Audi Electronics Venture GmbH) |
Bayesian Optimization for Deep Learning Based Object Detection | 06/2019 |
|
Deepti Joshi (Master thesis – co-supervison w/ Robert Bosch Automotive Steering GmbH) |
Verification of Reliability Prediction Models for Practical Suitability and Implementation of Improvements | 04/2019 |
|
Sakthivel Velumani (Master thesis – co-supervison w/ Intel Deutschland GmbH) |
Design and Implementation of Improved Decoding Algorithms for LDPC Convolutional Codes | 01/2019 |