Athanasios N. (Thanos) Nikolakopoulos, Ph.D.
Welcome
I'm Thanos. I am an Applied Scientist II at Amazon.com in Seattle, Washington, USA.
I am a Computer Science Researcher focusing on Machine Learning/Artificial Intelligence. My research interests span the areas of Natural Language Understanding, Graph Learning & Network Science, Recommender Systems, as well as Computational Biology.
Within these fields, my work focuses on developing novel algorithms and models as well as useful software tools to tackle challenging real-world problems. I have co-authored more than 30 papers, published in prestigious international conferences and journals (including ICML, KDD, RecSys, WSDM, IEEE TSP). I am an invited Program Commitee member on several premier international conferences focusing on AI/ML.
I have been honored to be the (co-) recipient of 3 Best Paper Awards together with my amazing collaborators, as well as an NSF awarded research grant.
Aside from my scientific work, I am a classical guitarist. In my teen years, I have performed in numerous international guitar festivals and concert halls, and have won 1st and 2nd prize in National classical guitar competitions. Nowadays, I play primarily for my wonderful wife Maria and our friends :)
Selected Publications
Projection techniques to update the truncated SVD of evolving matrices with applications
Vasileios Kalantzis, Georgios Kollias, Shashanka Ubaru, Athanasios N. Nikolakopoulos, Lior Horesh, Kenneth Clarkson Proceedings of the 38th International Conference on Machine Learning, 38th International Conference on Machine Learning (ICML 2021) PMLR 139:5236-5246, 2021. paper pdf | |
Best Paper Award |
Graph-based Recommendation with Personalized Diffusions
Athanasios N. Nikolakopoulos, Dimitris Berberidis, George Karypis, Georgios B. Giannakis MLG @ ACM KDD 2019 paper pdf code data slides |
RecWalk: Nearly Uncoupled Random Walks for Top-N Recommendation
Athanasios N. Nikolakopoulos, George Karypis ACM WSDM 2019 (Acceptance rate: 16.4%) Boosting Item-based Collaborative Filtering via Nearly Uncoupled Random Walks Athanasios N. Nikolakopoulos, George Karypis ACM Transactions on Knowledge Discovery from Data 14, 6, Article 64 (September 2020), 27 pages Conference paper pdf Journal paper pdf code data slides |
Best Paper Award |
Adaptive Diffusions for Scalable Learning over Graphs
Dimitris Berberidis, Athanasios N. Nikolakopoulos, Georgios B. Giannakis IEEE Transactions on Signal Processing, vol. 67, no. 5, pp. 1307-1321, March1, 2019. MLG @ ACM KDD 2018 paper pdf code data slides |
Best Paper Award |
EigenRec: Generalizing PureSVD for Effective and Efficient Top-N
Recommendations
Athanasios N. Nikolakopoulos, Vassilis Kalantzis, Efstratios Gallopoulos, John D. Garofalakis Knowledge and Information Systems (KAIS) May 2018 IEEE Int. Conference on Big Knowledge 2017 paper pdf code slides |
Elements and evolutionary determinants of genomic divergence between paired primary and metastatic tumors
Ruping Sun, Athanasios N. Nikolakopoulos PLOS Computational Biology, March 2021 paper pdf |
|
Random Surfing on Multipartite Graphs
Athanasios N. Nikolakopoulos, Antonia Korba and John D. Garofalakis IEEE BIGDATA 2016 paper pdf code data slides |
|
Hierarchical Itemspace Rank: Exploiting Hierarchy to Alleviate Sparsity in Ranking-based
Recommendation
Athanasios N. Nikolakopoulos, Marianna Kouneli and John D. Garofalakis Journal Neurocomputing Sep.2015, Elsevier paper pdf |
|
NCDREC: A Decomposability Inspired Framework for Top-N Recommendation
Athanasios N. Nikolakopoulos and John D. Garofalakis ACM/IEEE Web Intelligence 2014 paper pdf slides |
|
NCDawareRank: a Novel Ranking Method that Exploits the Decomposable Structure of the
Web
Athanasios N. Nikolakopoulos and John D. Garofalakis ACM WSDM 2013 paper pdf slides |
News
- September '21: I was invited to participate in a Panel Discussion at ACM RecSys 2021.
- July '21: I was invited to serve in the Scientific Program Commitees of ACM WWW 2022 and ACM WSDM 2022 and SIAM SDM 2022.
- June '21: Our paper "Projection techniques to update the truncated SVD of evolving matrices with applications", by Vasileios Kalantzis, Georgios Kollias, Shashanka Ubaru, Athanasios N. Nikolakopoulos, Lior Horesh, Kenneth Clarkson Proceedings of the 38th International Conference on Machine Learning, PMLR 139:5236-5246, 2021. has been accepted for publication, and was presented at the International Conference ICML 2021.
- June '21: I am starting as an Applied Scientist II, with Amazon.com in Seattle.
- March '21: The paper "Elements and evolutionary determinants of genomic divergence between paired primary and metastatic tumors," by Ruping, Sun and Athanasios N. Nikolakopoulos PLOS Computational Biology is available [here]
- October '20: The paper "Boosting Item-based Collaborative Filtering via Nearly Uncoupled Random Walks," by Athanasios N. Nikolakopoulos and George Karypis ACM Transactions on Knowledge Discovery from Data is available by the [ ACM Digital Library].
- February '20: I was invited to serve in the Program Commitees of ACM KDD 2020 and ACM RecSys 2020.
- August '19: We got an NSF Award (Co-PI) for our research project "CIF: Medium: Adaptive Diffusions for Scalable and Robust Learning over Graphs,". [more]
- August '19: The paper "Graph-Based Recommendation with Personalized Diffusions," by Athanasios N. Nikolakopoulos, Dimitris Berberidis, George Karypis, and Georgios B. Giannakis won the Best Paper Award at the ACM International Conference on Knowledge Discovery and Data mining 2019 (Mining and Learning with Graphs Workshop).
- July '19: Our paper "Personalized Diffusions for Top-N Recommendations," which was accepted for publication as a long paper in the ACM International Conference on Recommender Systems RecSys'2019 in Copenhagen Denmark.
- July '19: I was invited to serve in the Program Commitee of IEEE/ACM/ASA DSAA 2019.
- June '19: Our paper "Graph-Based Recommendations with Personalized Diffusions," was selected for oral presentation in the MLG@KDD'2019.
- February '19: I was invited to serve in the Program Commitees of ACM KDD 2019 and ACM RecSys 2019.
- February '19: I will be attending the ACM International Conference on Web Search and Data Mining WSDM'2019 in Melbourne Australia, where I will be presenting the paper "RecWalk: Nearly Uncoupled Random Walks for Top-N Recommendation"
- August '18: The paper "Adaptive Diffusions for Scalable Learning over Graphs," by Dimitris Berberidis, Athanasios N. Nikolakopoulos, and Georgios B. Giannakis won the Best Paper Award at the ACM International Conference on Knowledge Discovery and Data mining 2018 (Mining and Learning with Graphs Workshop). [more]
- August '17: The paper "Factored Proximity Models for Top-N Recommendations," by Athanasios N. Nikolakopoulos, Vassilis Kalantzis, Efstratios Gallopoulos and John D. Garofalakis won the Best Paper Award at the IEEE International Conference on Big Knowledge (IEEE ICBK’17). [more]
Professional Activities
Program Committee Member: ACM KDD 2019-2021, ACM RecSys 2019-2021, ACM WSDM 2022, The Web Conf (WWW) 2022, SIAM SDM 2018, 2022, PAKDD 2019-2018, IEEE DSAA 2019-2021. |
Journal Reviewer: IEEE TKDE, IEEE TSP, PLoS ONE, Data Mining and Knowledge Discovery, Neurocomputing Bioinformatics. |
Education
PhD in Computer Engineering and Informatics
- PhD Dissertation: Ranking Under Near Decomposability [pdf]
- Advising Commitee: Prof. Garofalakis (supervisor), Prof. Gallopoulos, Assoc.Prof. Nikoletseas Computer Engineering and Informatics Department, University of Patras, Greece.
MSc in Computer Science and Technology
- Master's Thesis (in Greek): Nearly Completely Decomposable Stochastic Systems and Applications,
Computer Engineering and Informatics Department, University of Patras, June 2011
Engineering Diploma (5-year program, M.Eng.) in Computer Engineering and Informatics
- Diploma Thesis (in Greek): An Application of the Theory of Decomposability to a Computer System
Performance Evaluation Problem,
Computer Engineering and Informatics Department, University of Patras, June 2010