|
Morteza Haghir Chehreghani
Associate Professor |
|
Since 2018, I am an Associate Professor in Machine Learning and Artificial Intelligence at the
Department
of Computer Science and Engineering, Chalmers University of
Technology. I received my Ph.D. from ETH
Zurich (Switzerland) in Computer Science
in 2014, where I worked at the Machine
Learning Institute under the supervision of Prof.
Joachim M. Buhmann. After the Ph.D., I spent about four years at Xerox
Research Centre Europe (later called Naver
Labs Europe) as a researcher (Staff Research Scientist I, and then Staff
Research Scientist II). Artificial Intelligence,
Machine Learning and Data Science constitute my general research interests. In particular, I am interested in Under-Supervised Machine
Learning and Decision Making, i.e., the AI-based decision-making problems that need to deal with the lack of sufficient available information (e.g., labelled data, decision
parameters, uncertainty and beyond). I study these problems in the contexts of Sequential
Decision Making (Interactive Machine
Learning) and Unsupervised Learning: § [current and future research] Interactive Machine Learning and Decision Making o
Active learning and active decision making o Online learning
(Bandits) o
(Deep) Reinforcement learning o
Bayesian optimization o Deep learning for
reinforcement/online/active learning and decision making § [more in the
past] Unsupervised Learning o
Cluster modelling o
Unsupervised
representation learning o
Learning
theory and generalization performance of unsupervised learning In all the problems, aspects such as generality, computational
efficiency, transparency and explainability are taken into account. My research has been applied to several
real-world problems in domains such as transport, life science, energy, automation, recommendation, and decision
support systems. Currently, I work with the following PhD
students and postdocs. o
Niklas
Åkerblom (industrial PhD student, Volvo Cars): Online learning of
energy-efficient navigation for electric vehicles o
Linus
Aronsson (academic PhD student): Active learning for deep neural network
models o
Ebrahim
Balouji (postdoc, in collaboration with E2, M2 and CEVT): Trip planning
and prediction with deep reinforcement learning o
Hampus
Gummesson Svensson (industrial PhD student, AstraZenece): Online/active
learning and exploration for automated drug discovery o
Fazeleh
S. Hoseini, (academic PhD student): Online learning of minimax distance
measures o
Arman
Rahbar (academic PhD student): Under-supervised representation learning · Advanced Topics in
Machine Learning (reinforcement learning, online learning, active learning,
and relevant deep learning models), Spring 2022. · Algorithms for
Machine Learning and Inference, Spring 2022. · Advanced Topics in
Machine Learning (reinforcement learning, online learning, active learning,
and relevant deep learning models), Spring 2021. ·
Algorithms for Machine Learning and Inference,
Spring 2021. ·
Algorithms for Machine Learning and Inference,
Spring 2020. ·
Statistical Methods for
Data Science, with Magnus V. Persson, Fall 2019. ·
Theoretical Foundations of
Machine Learning (PhD course), with Devdatt Dubhashi, Fredrik Johansson, Ashkan
Panahi, Fall 2019. ·
An intense course on
Machine Learning to the Chalmers CSE faculty, June 2019. ·
Algorithms for Machine Learning and Inference,
Spring 2019. ·
Statistical Methods for
Data Science, with Richard Johansson, Fall 2018. ·
Algorithms for Machine Learning and Inference,
with Devdatt Dubhashi, Spring 2018. ·
ML-HW CoDesign (Efficient
Deep Learning, PhD course), with Fredrik Dahlgren, Devdatt Dubhashi, Olof
Mogren, Miquel Pericas, Pedro Petersen, Ioannis Sourdis and Per Stenström,
Spring 2018.
Department of Computer Science and Engineering SE-412 96 Göteborg Sweden https://www.chalmers.se/en/staff/Pages/haghir.aspx
* Pre-prints and the under-review submissions can be found in the DBLP profile. [1] Niklas
Åkerblom, Fazeleh Sadat Hoseini, and Morteza
Haghir Chehreghani,
“Online Learning of Network Bottlenecks via Minimax Paths”,
Machine Learning (MLJ), subject to minor
revision, 2022. [2] Victor Eberstein, Jonas Sjöblom, Nikolce
Murgovski, and Morteza
Haghir Chehreghani, “A Unified Framework for Online Trip
Destination Prediction”, Machine Learning (MLJ), 2022. [3] Morteza Haghir Chehreghani, “Shift of Pairwise Similarities for Data Clustering”,
Machine Learning (MLJ), 2022. [4] Sanna Jarl, Linus Aronsson,
Sadegh Rahrovani, and Morteza Haghir
Chehreghani, “Active Learning
of Driving Scenario Trajectories”, Engineering
Applications of Artificial Intelligence, 2022. [5] Fazeleh Sadat Hoseini, and Morteza Haghir Chehreghani, “Memory-Efficient Minimax Distance
Measures”, 26th Pacific-Asia Conference on Knowledge Discovery and Data
Mining (PAKDD), 2022. [6] John Daniel Bossér, Erik Sörstadius, and Morteza Haghir Chehreghani, “Model-Centric and Data-Centric Aspects of Active Learning for
Deep Neural Networks”, IEEE International Conference on Big Data (IEEE
BigData), 2021. [7] Masoud Malek, Mostafa H. Chehreghani, Ehsan Nazerfard, and Morteza Haghir Chehreghani, “Shallow Node Representation Learning using Centrality Indices”,
IEEE International Conference on Big Data (IEEE BigData), 2021. [8] Fazeleh Hoseini, Sadegh Rahrovani, and Morteza Haghir Chehreghani, “Vehicle Motion Trajectories Clustering
Via Embedding Transitive Relations”, 24rd IEEE International
Intelligent Transportation Systems Conference (IEEE ITSC), 2021. [9] Sanna Jarl, Julia
Wennerblom, Maria Svedlund, Sadegh Rahrovani, and Morteza Haghir Chehreghani, “Annotation of Traffic Scenarios for AD Verification Using Active
Learning”, 6th International Symposium on Future Active
Safety Technology Toward zero traffic accidents (FAST-zero), 2021. [10] Morteza
Haghir Chehreghani,
“Reliable Agglomerative Clustering”, International
Joint Conference on Neural Networks (IJCNN), 2021. [11] Morteza
Haghir Chehreghani,
“Unsupervised Representation Learning with Minimax
Distance Measures”, Machine Learning
(MLJ), 109 (11), 2063-2097, 2020. [12] Morteza
Haghir Chehreghani, Mostafa H. Chehreghani, “Learning Representations from Dendrograms”, Machine Learning
(MLJ), 109 (9): 1779–1802, 2020. [13] Ashkan
Panahi, Morteza Haghir Chehreghani, Devdatt Dubhashi, “Accelerated Proximal Incremental Algorithm
Schemes for Non-Strongly Convex Functions”, Theoretical Computer Science, 812: 203-213, 2020. [14] Niklas
Åkerblom, Yuxin Chen, and Morteza Haghir
Chehreghani,
“An Online Learning Framework for Energy-Efficient
Navigation of Electric Vehicles”, 29th International
Joint Conference on Artificial Intelligence (IJCAI), 2020. [15] Andreas
Demetriou, Henrik Alfsvåg, Sadegh Rahrovani, and Morteza
Haghir Chehreghani,
“Generation of Driving Scenario Trajectories
with Generative Adversarial Networks”, 23rd IEEE International Intelligent Transportation Systems
Conference (IEEE ITSC), 2020. [16] Erik
Thiel, Morteza Haghir Chehreghani, and Devdatt Dubhashi, “A
Non-convex Optimization Approach to Correlation Clustering”, Thirty-Third
AAAI Conference on Artificial Intelligence (AAAI), 2019. [17] Claes
Strannegård, Herman Carlström, Niklas Engsner, Fredrik Mäkeläinen, Filip
Slottner Seholm, and Morteza Haghir Chehreghani, “Lifelong Learning Starting from Zero”, The 12th annual conference on Artificial
General Intelligence (AGI), 2019. [18] Claes
Strannegård, Herman Carlström, Niklas Engsner, Fredrik Mäkeläinen, Filip
Slottner Seholm, and Morteza Haghir Chehreghani, “Dynamic Deep Learning”, The 31st annual workshop of the Swedish Artificial
Intelligence Society (SAIS), 2019. [19] Mostafa
H. Chehreghani, and Morteza Haghir Chehreghani, “Efficient
Context-Aware K-Nearest Neighbor Search”, 40th European Conference on
Information Retrieval (ECIR), 2018. [long paper] [20] Yuxin
Chen, Jean-Michel Renders, Morteza Haghir Chehreghani, and
Andreas Krause, “Efficient Online Learning for Optimizing Value of
Information: Theory and
Application to Interactive Troubleshooting”, Thirty-Third Conference on Uncertainty in Artificial
Intelligence (UAI), 2017. [21] Morteza
Haghir Chehreghani,
“Classification with Minimax Distance Measures”, Thirty-First AAAI
Conference on Artificial Intelligence (AAAI), 2017. [22] Morteza
Haghir Chehreghani,
“Clustering by Shift”, IEEE International Conference on Data Mining
(ICDM), pp. 793-798, 2017. [23] Morteza
Haghir Chehreghani,
“Efficient Computation of Pairwise Minimax Distance Measures”, IEEE
International Conference on Data Mining (ICDM), pp. 799-804,
2017. [24] Morteza
Haghir Chehreghani,
“Feature-Oriented Analysis of User Profile Completion Problem”, 39th
European Conference on Information Retrieval (ECIR), 2017. [long paper] [25] Morteza
Haghir Chehreghani,
“Adaptive Trajectory Analysis of Replicator Dynamics for
Data Clustering”, Machine Learning
(MLJ), 104 (2-3): 271-289, 2016. [26] Morteza
Haghir Chehreghani,
Mostafa H. Chehreghani, “Modeling Transitivity in Complex Networks”,
Thirty-Second Conference on Uncertainty in Artificial Intelligence (UAI),
2016. [27] Mostafa
H. Chehreghani, Morteza Haghir Chehreghani, “Transactional
Tree Mining”,
European Conference on Machine Learning and Principles and Practice of
Knowledge Discovery (ECML/PKDD), (1) 182-198, 2016. [28] Morteza
Haghir Chehreghani,
“K-Nearest Neighbor Search and Outlier Detection via Minimax Distances”,
SIAM International Conference on Data Mining (SDM), 2016. [29] Ana
Paula Oliveira, Sotiris Dimopoulos, Alberto Giovanni Busetto, Stefan
Christen, Reinhard Dechant, Laura Falter, Morteza Haghir Chehreghani,
Szymon Jozefczuk, Christina Ludwig, Florian Rudroff, Juliane Schulz,
Alexandre Soulard, Daniele Stracka, Ruedi Aebersold, Joachim M. Buhmann,
Michael N. Hall, Matthias Peter, Uwe Sauer, and Jörg Stelling, “Inferring causal metabolic signals that regulate the
dynamic TORC1-dependent transcriptome”,
Molecular Systems Biology,11(4), 2015. [30] Ludwig
M. Busse, Morteza Haghir Chehreghani, Joachim M. Buhmann, “Approximate
Sorting”, 35th German Conference on
Pattern Recognition (GCPR), 142-152, 2013. [31] Morteza
Haghir Chehreghani,
Ludwig M. Busse, Joachim M. Buhmann, “Information Content in Clustering
Algorithms”, The Deutsche ArbeitsGemeinschaft Statistik Conference
(DAGStat), 2013. [32] Morteza
Haghir Chehreghani, Alberto G. Busetto, Joachim M.
Buhmann, “Information Theoretic Model Validation for Spectral Clustering”,
International Conference on Artificial Intelligence and Statistics (AISTATS),
pp. 495-503, 2012. [33] Morteza
Haghir Chehreghani,
Mostafa H. Chehreghani, and Hassan Abolhassani, “Probabilistic Heuristics for Hierarchical Web Data
Clustering”, Computational
Intelligence, 28(2): 209-233, 2012. [34] Ludwig
M. Busse, Morteza Haghir Chehreghani, Joachim M. Buhmann, “The
Information Content in Sorting Algorithms”, International Symposium on
Information Theory (ISIT), 2746-2750, 2012. [35] Alberto
G. Busetto, Morteza Haghir Chehreghani, Joachim M. Buhmann, “Approximate
Set Coding for Information Theoretic Model
Validation”,
The Fifth Workshop on Information Theoretic Methods in Science and
Engineering (WITMSE), pp. 15-18, 2012. [36] Joachim
M. Buhmann, Morteza Haghir Chehreghani, Mario Frank, and Andreas P.
Streich, “Information Theoretic Model Selection for Pattern
Analysis”, Journal of Machine
Learning Research (JMLR), extended from ICML workshop on Unsupervised
and Transfer Learning, 27:51–64, 2012. [37] K. H.
Brodersen, Z. Lin, A. Gupta, W.D. Penny, A.P. Leff, Morteza Haghir
Chehreghani, Alberto G. Busetto, Joachim M. Buhmann, K.E. Stephan, “Model-based
clustering using generative embedding”, Human Brain Mapping, 2012. [Oral
presentation] [recipient of Trainee Abstract
Award]. [38] Mostafa
H. Chehreghani, Morteza Haghir Chehreghani, Caro Lucas, and Masoud
Rahgozar, “OInduced: An Efficient Algorithm for Mining Induced
Patterns from Rooted Ordered Trees”,
IEEE Transactions on Systems, Man, and Cybernetics, Part A 41(5): 1013-1025,
2011. [39] Ludwig
M. Busse, Morteza Haghir Chehreghani, Joachim M. Buhmann, “Approximate
Sorting of Preference Data”, NIPS workshop on
Choice Models and Preference Learning, 2011. [selected
for NIPS workshop contributed talk] [40] Morteza
Haghir Chehreghani, Alberto G. Busetto, Joachim M. Buhmann, “Cluster
Model Validation by Maximizing Approximation Capacity”,
NIPS workshop on New Frontiers in Model Order Selection, 2011. [received NIPS travel grant] [41] Mario
Frank, Morteza Haghir Chehreghani, and Joachim M. Buhmann, “The
Minimum Transfer Cost Principle for Model-Order Selection”, European Conference
on Machine Learning and Knowledge Discovery in Databases (ECML/PKDD),
(1) 423-438, 2011. [42] Morteza
Haghir Chehreghani,
Hassan Abolhassani, and Mostafa H. Chehreghani, “Density
link-based methods for clustering web pages”,
Decision Support Systems, 47 (4): 374-382, 2009. [43] Mostafa
H. Chehreghani, Morteza Haghir Chehreghani, Masoud Rahgozar, Caro
Lucas, and Euhanna Ghadimi, “Efficient Rule Based
Structural Algorithms for Classification of Tree-Structured Data”, Intelligent Data Analysis, 13 (1):
165-188, 2009. [44] Morteza
Haghir Chehreghani,
Hassan Abolhassani, and Mostafa H. Chehreghani, “Improving
Density Based Methods for Hierarchical Clustering of the Web Pages”, Data and Knowledge
Engineering, 67 (1): 30-50, 2008. [45] Mehrdad
Mahdavi, Morteza Haghir Chehreghani, Hassan Abolhassani, and Rana Forsati, “Novel Meta-Heuristic Algorithms for
Clustering Web Documents”, Applied
Mathematics and Computation, 201 (1-2): 441–451, 2008. [46] Mohammad
Rahimi, Morteza Haghir Chehreghani, Hassan Abolhassani, “Improving
Tree Structures to Find Dense Clusters from Web Documents”,
International Conference on Information and Knowledge Technology (IKT), 2008. [47] Mostafa
H. Chehreghani, Masoud Rahgozar, Caro Lucas, and Morteza Haghir
Chehreghani, “A Heuristic
Algorithm for Clustering Rooted Ordered Trees”,
Intelligent Data Analysis, 11 (4): 355-376, 2007. [48] Morteza
Haghir Chehreghani,
and Hassan Abolhassani, “Attaining Higher
Quality for Density Based Algorithms”, International
Conference on Web Reasoning and Rule Systems (RR), LNCS, 329–338, 2007. [49] Mostafa
H. Chehreghani, Masoud Rahgozar, Caro Lucas, Morteza Haghir Chehreghani,
“Mining Maximal Embedded Unordered Tree
Patterns”, IEEE Symposium on Computational
Intelligence and Data Mining (CIDM), 437-443, 2007. [50] Morteza
Haghir Chehreghani,
Hassan Abolhassani, “H-BayesClust: A new Hierarchical Clustering based on
Bayesian Networks”, International Conference
on Advanced Data Mining and Applications (ADMA), LNCS, pp. 616–624, 2007. [51] Morteza
Haghir Chehreghani, Mohammad Rahmati, “Design and
Implementation of a 2-D Barcode based on Data Matrix Format”,
15th Iranian Conference on Electrical Engineering, 2006. [1] Morteza
Haghir Chehreghani,
“System and method for performing k-nearest neighbor
search based on minimax distance measure and efficient outlier detection”,
Patent number: 10073887, Date of Patent:
September 11, 2018, https://patents.justia.com/patent/10073887 [Type: Grant] [2] Morteza Haghir Chehreghani, “Efficient Calculation of All-Pair Path-based
Distance Measures”, Patent number: 9805138,
Date of Patent: October 31, 2017, https://patents.justia.com/patent/9805138 [Type: Grant] [3] Morteza Haghir Chehreghani and Yuxin Chen, “Method of Trip
Prediction by Leveraging Trip Histories from Neighboring Users”, Publication number: 20180012141,
Publication date: January 11, 2018, https://patents.justia.com/patent/20180012141 [Type: Application] [4] Morteza Haghir Chehreghani, “Adaptive Trajectory Analysis of Replicator Dynamics
for Data Clustering”, Publication number: 20160179923, Publication date: June 23, 2016, https://patents.justia.com/patent/20160179923 [Type: Application] ·
An efficient method for
correlation clustering with fixed and variable number of clusters ·
Efficient computation
of Minimax distances |
|