May 1st 2023. I start my position at Zalando SE as a senior principal scientist in Berlin, Germany. Zalando is a leading European online platform for fashion and lifestyle with more than 50 million active customers across 25 countries. More about Zalando can be found at corporate.zalando.com.
March 11th 2023. My patent, Artificial Intelligence System Using Deep Neural Networks for Pairwise Character-Level Text Analysis and Recommendations, US Patent 11,599,927, is granted by the US Patent Office. This was a joint invention by Valentin Flunkert and me.
June 21st 2022. My patent, Artificial Intelligence System for Generating Network-Accessible Recommendations with Explanatory Metadata, US Patent 11,367,117, is granted by the US Patent Office. The patent application was filed when I worked with the Amazon Gifting team to help customers find their ideal gifts.
January 28th 2022. I have been involved in the peer review process for the following ML/NLP conferences in 2021: ACL (area chair), AAAI (senior PC), NIPS, UAI, AISTATS, ICLR, ACML, EMNLP, NAACL-HLT.
June 3rd 2021. I become an Amazon Bar Raiser. A Bar Raiser is a skilled interviewer on the hiring committee who acts as a steward of the interview process. It has three responsibilities: Assess candidates for the specific role and for long-term success; Make sure there is an accurate and fair assessment of the candidate with every member of the interview loop; Guide and help with the preparation of the interview.
February 25th 2021. My patent, Artificial Intelligence System for Automated Generation of Realistic Question and Answer Pairs, US Patent 10,929,392, is granted by the US Patent Office. The patent application was filed when I worked on improving customer engagement on Amazon product detail pages.
January 4th 2021. I am serving as an area chair for the Question Answering track at ACL-IJCNLP 2021, The Joint Conference of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing. The conference will be held in Bangkok, Thailand, during August 1-6, 2021.
October 6th 2020. My patent, Extracting Keywords from a Document, US Patent 10,796,094, is granted by the US Patent Office. This patent is a continuation of US Patent 10,387,568, granted in August 2019.
August 7th 2020. I am serving as a senior program committee member (SPC) for AAAI 2021, the 35th AAAI Conference on Artificial Intelligence. AAAI 2021 is held on February 2-9 as a virtual conference.
February 28th 2020. I am serving as an area chair at EMNLP 2020, the 2020 Conference on Empirical Methods in Natural Language Processing, for the Question Answering track. EMNLP 2020 is held in Dominican Republic from November 8th to 12th.
February 11th 2020. My patent, Document Content Analysis Based on Topic Modeling, US Patent 10,558,657, is granted by the US Patent Office.
August 20th 2019. My patent, Extracting Keywords from a Document, US Patent 10,387,568, is granted by the US Patent Office.
June 3rd 2019. I am serving as an area chair at EMNLP-IJCNLP 2019, the 2019 Conference on Empirical Methods in Natural Language Processing and 9th International Joint Conference on Natural Language Processing, for the track Textual Inference and Other Areas of Semantics. EMNLP-IJCNLP 2019 is held in Hong Kong from November 3rd through November 7th.
April 10th 2019. My patent, Document Content Analysis Based on Topic Modeling, US Patent 10,255,283, is granted by the US Patent Office.
April 3rd 2019. My patent, Generating a Set of Representative Items Using a Maximum-Set-Coverage Selection Strategy, US Patent 10,248,712, is granted by the US Patent Office.
January 1st 2019. I am serving as a senior program committee member (SPC) for IJCAI 2019, the 28th International Joint Conference on Artificial Intelligence. IJCAI 2019 is held on August 10-16 in Macao, China. Please consider submitting your work!
October 30th 2018. Two of my patents are granted by the US Patent Office: Generating a Set of Representative Items Using a Clustering-Selection Strategy, US Patent 10,114,885, and Generating a Set of Representative Items Using a Dynamic Selection Strategy, US Patent 10,114,887.
May 31st 2018. Our paper Multiplicative Tree-Structured Long Short-Term Memory Networks for Semantic Representations is accepted for a long oral presentation at *SEM 2018, the 7th Joint Conference on Lexical and Computational Semantics. This is a joint work with Nam Khanh Tran, while he was interning with me in 2017. *SEM 2018 takes place in New Orleans, US, co-located with NAACL and SemEval.
- September 5th 2017. I give a talk at Data Science Luxembourg on Salience Rank, our new keyphrase extraction algorithm published at ACL 2017. This is a regular gathering of Luxembourg data science community and there were around 100 participants this time.
- March 1st 2017. If you are a PhD student with strong machine learning or natural language processing background, feel free to contact me for an internship at Amazon Development Center in Berlin, Germany. Candidates with interests in deep text understanding and semantics are especially encouraged to apply. (Update: The position for summer 2017 is filled.)
- December 10th 2016. I attend COLING 2016, the 26th International Conference on Computational Linguistics, in Osaka, Japan. The conference takes place on December 11-16.
- November 7th 2016. I attend DA2PL 2016, From Multiple Criteria Decision Aid to Preference Learning. DA2PL 2016 is a EURO Mini Conference that aims to bring together researchers from decision analysis and machine learning. This year DA2PL is held at the Paderborn University in Germany, November 7-8.
- August 8th 2016. I attend ACL 2016, the 54th annual meeting of the Association for Computational Linguistics. This year ACL is held at the Humboldt University in Berlin, Germany, August 7-12.
- October 5th 2015. I attend the Yandex School of Data Analysis Conference in Berlin, Germany. In this conference, Prof. Vladimir Vapnik will discuss his recent work on Intelligent Learning, a new theoretical framework of machine learning.
- July 29th 2015. My paper, Label Ranking by Learning Pairwise Preference, co-authored with Eyke Hüllermeier, Johannes Fürnkranz, Klaus Brinker, receives the Prominent Paper Award from the Artificial Intelligence journal. This award recognizes outstanding papers published not more than seven years ago in the AI journal that are exceptional in their significance and impact. The citation for our award:
"This paper is a key paper in the area of preference learning. It studies the problem of label ranking, which is concerned with learning a mapping from instances to rankings over a finite number of labels. The authors introduce the Ranking by Pairwise Comparison algorithm (RPC), which first induces a binary preference relation and then uses this relation to derive a ranking. The paper contains appealing theoretical results (that RPC can minimize different loss functions) as well as empirical results (that RPC is competitive in terms of accuracy and superior in terms of efficiency). The paper shows the elegance and power of a natural and intuitively appealing approach. It has been influential in the field of preferences and preference learning."
More information about this award and previous winners can be found here. The award ceremony is held at the opening of IJCAI 2015 in Buenos Aires, Argentina.
- November 10th 2014. I will be reviewing for ICML 2015, the 32nd International Conference on Machine Learning. The conference takes place in Lille, France from July 6th to 11th.
- November 3rd 2014. I attend CIKM 2014, the 23rd ACM International Conference on Information and Knowledge Management. The 5-day conference is hosted in Shanghai, China.
- April 23rd 2014. I will be reviewing for NIPS 2014, the 28th Annual Conference on Neural Information Processing Systems. The conference takes place in Montreal, Canada between December 8th and 13th.
- March 25th 2014. Invited by Prof. Krzysztof Dembczyński and Prof. Roman Słowiński, I give a talk on Preference Learning using Statistical Methods for Label Ranking at Poznan University of Technology (Politechnika Poznańska, aka PUT). PUT is known as one of the best technical universities in Poland and has an outstanding computer science department.
- March 2nd 2014. I attend the Dagstuhl Seminar on Preference Learning, which takes place in Schloss Dagstuhl from March 2nd to 7th. This event is organized by Johannes Fürnkranz, Eyke Hüllermeier, Cynthia Rudin, and Roman Słowiński. A group picture of the participants is available here.
- July 11th 2013. Our paper Preference-Based CBR: A Search-Based Problem Solving Framework wins the best paper award at the 21st International Conference on Case-Based Reasoning (ICCBR 2013). This paper is a joint work by Amira Abdel-Aziz, Marc Strickert, Eyke Hüllermeier, and me.
- July 1st 2013. I start my position at Amazon.com as a machine learning scientist in Berlin, Germany and Seattle, USA. More information about machine learning at Amazon can be found at amazon.jobs.
- May 22nd 2013. Our paper On Label Dependence and Loss Minimization in Multi-label Classification, published at Machine Learning journal, is listed in Notable Computing Books and Articles 2012 by ACM Computing Reviews (pdf). This paper is written by Krzysztof Dembczyński, Willem Waegeman, Eyke Hüllermeier, and me.
- March 19th 2013. I am invited to the program committee of ECMLPKDD 2013. ECMLPKDD is the premier European machine learning and data mining conference and builds upon a very successful series of 23 ECML and 16 PKDD conferences, which have been jointly organized for the past 12 years. It takes place this year in Prague from September 23th to 27th. In addition to the traditional proceedings track, I am also involved with the reviewing process of the journal track, which is new to ECMLPKDD.
- January 25th 2013. I am invited to the program committee of IJCAI 2013, the 23rd International Joint Conference on Artificial Intelligence, to be held this August in Beijing, China. Held biennially in odd-numbered years since 1969, IJCAI is the main international gathering of researchers in AI.
- September 17th 2012. SUM 2012 conference takes place today in Marburg, Germany. SUM stands for Scalable Uncertainty Management and is an annual conference that was launched in 2007 with the goal to exploit and strengthen the connection between the Artificial Intelligence and Database communities. It aims at bringing together all those researchers interested in the management of massive amounts of uncertain, incomplete or inconsistent information. This year's SUM is co-organized by our group.
- July 30th 2012. The Machine Learning journal special issue of selected papers from ECMLPKDD 2011 is now available. Out of 599 submissions, two of my papers are selected among the other five. Those two papers are: Towards Preference-Based Reinforcement Learning, a joint work with Johannes Fürnkranz, Eyke Hüllermeier, Sang-Hyeun Park, and Learning Monotone Nonlinear Models using the Choquet Integral, a joint work with Ali Fallah Tehrani, Krzysztof Dembczyński, Eyke Hüllermeier. Read the editors' introduction to the special issue here.
- May 25th 2012. I have successfully defended my PhD thesis Label Ranking with Probabilistic Models and have been conferred with the doctoral degree Dr. rer. nat (the doctor of natural sciences) from the Faculty of Mathematics and Computer Science, Philipps University Marburg. The members of my examination committee are: Prof. Eyke Hüllermeier, Prof. Johannes Fürnkranz, Prof. Bernhard Seeger, and Prof. Volkmar Welker.
- April 6th 2012. Our Kebi team wins the 2nd place at JRS 2012 Data Mining Competition. This competition is a special event of Joint Rough Sets Symposium that takes place in Chengdu, China, August 17-20, 2012. The task is to predict topical classification of scientific publications in the field of bio-medicine. The members of the Kebi team are Krzysztof Dembczyński, Eyke Hüllermeier, Adrian Jaroszewicz, Willem Waegeman, and me.
- February 16th 2012. My paper, Label Ranking by Learning Pairwise Preferences, appeared in Artificial Intelligence journal (vol. 172, pp. 1897-1916), is listed as one of the most cited Artificial Intelligence articles published since 2007 (link, pdf). This paper is a joint work with Eyke Hüllermeier, Johannes Fürnkranz, and Klaus Brinker.
- February 8th 2012. I am invited to the program committee of (ECMLPKDD 2012), the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, to be held this September in Bristol, UK. ECMLPKDD 2012 builds upon a series of 22 ECML and 15 PKDD conferences, which have been jointly organized for the past 11 years.
- February 1st 2012. I am invited to the program committee of the 1st international workshop on Knowledge Discovery and Data Mining Meets Linked Open Data (Know@LOD 2012), to be held at the 9th Extended Semantic Web Conference (ESWC 2012) in Kalamaki, Greece. This workshop is jointly organized by Johanna Völker, Heiko Paulheim, Jens Lehmann, and Mathias Niepert.
- January 24th 2012. My talk on "Label ranking with abstention: Predicting partial orders by thresholding probability distributions" at NIPS 2012 workshop Choice Models and Preference Learning is now available on VideoLectures.net:
- January 12th 2012. I am invited to the program committee of the 20th European Conference on Artificial Intelligence (ECAI 2012). ECAI is the leading conference on Artificial Intelligence in Europe, held every two years. It gives researchers from all over the world the possibility to identify important new trends and challenges in all subfields of Artificial Intelligence, and provides a major forum for potential users of innovative AI techniques. This year's conference is jointly organized by the European Coordination Committee for Artificial Intelligence (ECCAI), the French Association for Artificial Intelligence (AFIA), and Montpellier Laboratory for Informatics, Robotics, and Microelectronics (LIRMM).
- December 17th 2011. Among 4 of 15 accepted submissions, our work Label Ranking with Abstention: Predicting Partial Orders by Thresholding Probability Distributions is selected for an oral presentation at the NIPS workshop Choice Models and Preference Learning. This workshop is held at Sierra Nevada, Spain.
- November 5th 2011. I am kindly selected for a travel award for attending the 25th Annual Conference on Neural Information Processing Systems (NIPS 2011), to be held at Granada and Sierra Nevada, Spain. This year two of my papers appear at NIPS, An Exact Algorithm for F-Measure Maximization together with Krzysztof Dembczyński, Willem Waegeman, Eyke Hüllermeier (at main conference) and Label Ranking with Abstention: Predicting Partial Orders by Thresholding Probability Distributions together with Eyke Hüllermeier (at the Choice Models and Preference Learning workshop).
- November 3rd 2011. My talk on "Learning Monotone Nonlinear Models using the Choquet Integral" at ECMLPKDD 2011 and LWA 2011 are now available respectively on VideoLectures.net and YouTube:
- September 20th 2011. China Scholars Abroad magazine features an article about my research on the 259th issue (September 2011, Page 38). This article is written in Chinese.
- September 8th 2011. Both of our submissions to ECMLPKDD, Learning Monotone Nonlinear Models using the Choquet Integral and Preference-Based Policy Iteration: Leveraging Preference Learning for Reinforcement Learning, are invited to appear at the Machine Learning Journal special issue devoted to ECMLPKDD 2011. In total, 10 papers are invited, selected from 599 submissions.
- August 12th 2011. The paper Automated Feature Generation from Structured Knowledge is accepted at CIKM 2011 as a full paper and for an oral presentation. CIKM has a very tough selection process this year: Only 15% of the submissions are accepted as full papers. This paper is a joint work with Gjergji Kasneci, Thore Graepel, David Stern, Ralf Herbrich and is a result of my internship last year at Microsoft Research Cambridge.
- July 22nd 2011. Our paper Choquistic Regression: Generalizing Logistic Regression using the Choquet Integral (co-authored by Ali Fallah Tehrani, me, and Eyke Hüllermeier) is nominated for the Best Student Paper award at the EUSFLAT 2011 conference. The basic idea of our approach is to replace the linear function of predictor variables, which is used in logistic regression to model the log odds of the positive class, by the Choquet integral. Thus, it becomes possible to capture non-linear dependencies and interactions among predictor variables while preserving important properties of logistic regression. The paper will soon be available in the proceedings of EUSFLAT 2011. Together with the presentation slides, it will appear in my publication list as well.
- May 18th 2011. I attend the 2011 SIAM Conference on Optimization, OP11. Sponsored by SIAM (Society for Industrial and Applied Mathematics) Activity Group on Optimization, OP11 presents the latest research in theory, algorithms, and applications in optimization, with a particular emphasis on large-scale problems. This year it has featured a highly interesting session on the research of ranking algorithms.
- April 20th 2011. Together with Prof. Ilka Agricola and Prof. Rita Loogen, I am initiating the Consulting Service for Asian Students in the Faculty of Mathematics and Computer Science at Philipps University Marburg. The mission of this consulting service is to promote the academic success and holistic experience of Asian students in our faculty. There are about thirty to forty students from China and other Asian countries studying here. Most of them have left the family for the first time, started a new life in an unfamiliar place. We hope through this new service we can help them with various issues they encounter during the study and the personal life.
- March 6th 2011. I am recruited as a program committee member of ECMLPKDD 2011, the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases. ECMLPKDD is a leading conference in the field and a major European scientific event. It builds upon a series of 21 ECML and 14 PKDD conferences, which have been jointly organized for the past ten years. ECMLPKDD 2011 takes place in Athens, Greece from September 5th to 9th.
- December 7th 2010. I present a demo at Think Computer Science, an one day event co-organized by Microsoft Research Cambridge. Every year, for the past 6 years, a number of students are invited for an exciting day of talks, demos, and interactive sessions. The goal is to inspire and support students early on in their education, providing insight into the world of a computer scientist. The demo I present (called RipOff!) together with a tutorial attempts to show some interesting aspects in cooperative game theory. It is based on the research of my colleague Yoram Bachrach.
- November 4th 2010. I participate in the Royal Society seminar Computational Frontiers in Scientific Discovery in London. This seminar brings together the recipients of the Royal Society and Académie des Sciences Microsoft Award, with leading scientists at the forefront of their disciplines, ranging from astrophysics and climate prediction, to genome biology and human variation. It explores the similarities and differences between the computational approaches across different disciplines and highlight potential synergies, whereby one discipline can borrow from another to speed up the process of discovery.
- From October 14th 2010, I work as a research intern at Machine Learning and Perception Group, Microsoft Research Cambridge. The focus of my internship is to improve machine learning with large-scale ontological knowledge. This internship lasts for 3 months, during which I can be reached via email@example.com.
- September 9th 2010. I am recruited as a reviewer for Journal of Machine Learning Research, a leading journal in computer science focusing on machine learning. JMLR is an open-access journal. All the accepted papers are published electronically (ISSN 1533-7928). The paper volumes are published and sold by Microtome Publishing (ISSN 1532-4435).
- July 29th 2010. I receive the UNESCO Conference Grant at the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECMLPKDD 2010). This grant, sponsored by UNESCO Chair in Data Privacy, aims to offset the cost of attending ECMLPKDD 2010 for selected participants. ECMLPKDD 2010 builds upon a very successful series of 20 ECML and 13 PKDD conferences, which have been jointly organized for the past nine years. It has become the major European scientific event in these fields.
- It is my great honor to be named a winner of the 2009 Chinese Government Award for Outstanding Self-Financed Students Abroad. I am one of 37 winners in Germany to receive the 2009 award, which includes a $5,000 prize and a certificate. This award is presented by the China Scholarship Council annually to recognize the academic merit and research accomplishments of exceptional, self-financed (not government-financed) students from China who are pursuing doctoral degrees abroad. It is granted around the world across all disciplines. The award ceremony in Germany was held at the Embassy of China in Berlin on June 5th 2010, where I have given the award acceptance speech on behalf of all the award winners. Here are a selection of news reports from the Chinese media as well as news articles from the press office at Philipps University Marburg and Oberhessische Presse.
- May 23rd 2010. I am awarded with the Student Travel Scholarship at the 27th International Conference on Machine Learning (ICML 2010). The goals of this scholarship are to encourage Ph.D. students to exchange ideas and to give them a forum to communicate with other researchers in the field. Recipients of this scholarship are provided financial supports to subsidize travel, conference registration, housing, and other expenses for ICML 2010. Moreover, they have the chance to show their recent research in the poster sessions during the conference. The scholarship is sponsored by IBM, NSF, etc. Here is the list of all recipients.
- April 17th 2010. All of my three submissions, joint work with Krzysztof Dembczyński and Eyke Hüllermeier, to the 27th International Conference on Machine Learning (ICML 2010) are accepted! Those submissions are: Label Ranking Methods based on the Plackett-Luce Model, Graded Multi-Label Classification: The Ordinal Case, and Bayes Optimal Multi-Label Classification via Probabilistic Classifier Chains. Drafts of these papers will soon appear in my publication list. More about the reviewing process of ICML 2010 can be seen at a post of my personal blog.
- March 15th 2010. I am nominated for the "2009 Chinese Government Award for Outstanding Self-Financed Students Abroad". This award is presented by the China Scholarship Council (CSC) and aims to encourage research excellence and recognize the achievement among Chinese students studying abroad. It is granted across all disciplines. This year 37 Chinese students in Germany have been nominated. Here is the list of all nominees for 2009 (in Chinese).
- On November 10th 2009, I have given a tutorial on text classification at GTO IES Audit & Risk Management Department, Deutsche Bank AG. In this tutorial I have introduced some general concepts of statistical classification and demonstrate the text classification system I developed for Deutsche Bank.
- October 26th 2009. I am recruited as a reviewer for the 27th International Conference on Machine Learning (ICML 2010), the premier scientific conference in machine learning, pattern recognition, and related fields. ICML 2010 will be held in Haifa, Israel, June 21-24, 2010, and will be co-located with The 23rd International Conference on Learning Theory (COLT 2010). The call for paper can be found at here.
- My talk on Combining Instance-Based Learning and Logistic Regression for Multi-Label Classification at LWA 2009, TU Darmstadt is now available on YouTube:
- September 25th 2009. I am invited to give a talk on human and computer intelligence at Otto von Guericke University of Magdeburg. This event is organized by Association of Chinese Scholars and Students in Magdeburg e.V.
- I have been invited to attend the 4th Annual Google Test Automation Conference (GTAC 2009) in Zurich, Switzerland on September 21st and 22nd 2009, and have given a 5-minute talk on multi-label classification in the "Lightning Talk" session. This event was organized by Google Zurich office. The talk is now available on YouTube:
- Invited by Prof. Manfred Sommer, I participated in the Ernst & Young seminar hosted in our faculty on November 4th 2009. Ernst & Young is one of the largest professional services firms in the world and one of the Big Four auditors, along with PricewaterhouseCoopers, Deloitte, and KPMG. In this seminar, a throughout overview of the financial services that Ernst & Young offers was presented to the audience, together with some underlying statistical models applied at Ernst & Young.
- Video streams of my talks at ICML 2009 and ECMLPKDD 2009 are available at VideoLectures.net:
- On September 7th 2009 I was invited to the Round-Table dinner hosted by McKinsey & Company Inc. at the restaurant Zur Sonne, Marburg. McKinsey is a global management consulting firm that focuses on solving issues of concern to senior management and is widely recognized as a leader and one of the most prestigious firms in the management consulting industry. This Round-Table event was created to establish the relationship between (post-)graduate students at Philipps University Marburg and McKinsey. Twelve (post-)graduate students were invited. Among them, two were from the Faculty of Mathematics and Computer Science. During the dinner, the representatives from McKinsey have introduced the policy of the company, its long-term goal, and the working environment. This is also a get-to-know event for the consulting business at McKinsey.
- From August 3rd 2009, I have started working on a text mining project at GTO Deutsche Bank and thus appear at the campus only on Thursday and Friday. In case of need, you can reach me via email: firstname.lastname@example.org or telephone: +49-69-91067416 (from Monday till Wednesday between 9am to 5pm). The duration of this project is approximately 6 months.
- June 30th 2009. A paper from the KEBI lab, Combining Instance-Based Learning and Logistic Regression for Multi-Label Classification, co-authored with Eyke Hüllermeier, wins the "Machine Learning Journal Best Student Paper Award" at 2009 European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECMLPKDD 09). In this paper, we create a novel framework for multi-label classification that is able to explore the interdependencies between labels by combining nearest neighbor and logistic regression learning. This best paper award includes a €3,000 prize, a certificate, and a copy of Machine Learning Journal Special Issue for ECMLPKDD 2009 published by Springer.
- September 27th 2006. I participate in Symposium on Fuzzy Systems in Computer Science 2006 (FSCS 2006), invited by Prof. Eyke Hüllermeier. Prof. Lotfi Zadeh from UC Berkeley, the inventor of fuzzy logic, presents the keynote talk "A New Frontier in Computation: Computation with Information Described in Natural Language".