Rishabh Mehrotra

AI Researcher

Hi, i'm Rishabh Mehrotra!

Computer Scientist & Researcher

Research Scientist at Spotify Research, London. PhD in ML/IR from UCL. Interested in teaching machines to better understand, comprehend & support user needs.



  • [April, 2018] Gave the tutorial on Understanding User Needs & Tasks at WWW 2018, Lyon, France.
  • [Feb, 2018] Organized the 2018 Workshop on Learning from User Interactions at WSDM 2018, Los Angeles.
  • [Nov, 2017] Gave the tutorial on Understanding User Needs & Tasks at CIKM 2017, Singapore.
  • [Sept, 2017] Invited talk on Deep sequential models for Conversational Agents at PyData Delhi 2017.
  • [Aug, 2017] 2 papers (on deep sequential model & task embeddings) accepted at CIKM 2017, Singapore.
  • [June, 2017] Invited talk on Auditing Search Engines at K4All Workshop on Side-effects of Online Content Delivery, London, UK.
  • [June, 2017] Full paper accepted at ICTIR 2017.
  • [April, 2017] 2 full papers accepted at SIGIR 2017.
  • [Feb, 2017] Poster on Session identification for Intelligent Assistants accepted at WWW 2017 Posters track.
  • [Dec, 2016] Full paper on Auditing search engines for systemic bias accepted at WWW 2017.
  • [Nov, 2016] Gave a tutorial on Inferring User Tasks & Needs at Search Solutions 2016.
  • [July, 2016] Excited to be working on Cortana with Imed Zitouni & Ahmed Hassan at Microsoft Research, Redmond.
  • [June, 2016] Gave an invited talk on Digital Assistants, Proactive Recommendations & Search Tasks at Machine Learning meetup, Gurgaon.
  • [June, 2016] Paper accepted at ECML 2016.
  • [April, 2016] We're organizing the TREC 2016 Tasks Track. Consider participating!
  • Education

    • University College London Feb 2014 - Nov 2017

      Ph.D. in Computer Science

      Dept. of Computer Science

      Advisor: Dr. Emine Yilmaz

      Area: Machine Learning, Information Retrieval

    • BITS Pilani2008 - 2013

      B.E.(Hons.) Computer Science

      M.Sc.(Hons.) Mathematics

      Thesis: Sparsity & Structured Sparsity for Learning Coupled Representations from Multi-View Data

    Research Experience

    • Sept'16 June'16

      Microsoft Research, Redmond

      Research Intern, Measurement Sciences Team
      Mentors: Imed Zitouni, Ahmed Hassan
      Description: Developed terascale machine learning based solutions for understanding user interaction sequences for satisfaction prediction. Mining terabytes of logs for developing task based satisfaction metrics.

    • June'16 March'16

      Microsoft Research, New York

      Research Intern, Machine Learning Group
      Mentors: Fernando Diaz
      Description: Developed techniques for auditing search engines for demographic bias in performance.

    • Sept'15 July'15

      Microsoft Research, Redmond

      Research Intern, CLUES Group
      Mentors: Susan Dumais, Paul Bennett
      Description: Responsible for designing and developing terascale machine learning based solutions for cross-domain personalization. Mining terabytes of user logs to derive insights for personalization.

    • June'15 April'15

      Microsoft Bing, London

      Applied Scientist Intern, Related Searches Team
      Mentors: Abhishek Arun, Katja Hofmann
      Description: Developed techniques for counterfactual estimation of online metrics.

    • Dec'13 June'13

      Goldman Sachs, Bangalore

      Techology Analyst Reference Data Team
      Description: Developed Regression Testing framework for an event-driven SOA based framework.

    • Aug'12 May'12

      Google Summer of Code (GSoC)

      Xapian: search engine library
      Description: Implemented Learning to Rank algorithms for LETOR framework.

    • July'12 May'12

      NICTA, Australia

      Machine Learning Intern
      Mentors: Scott Sanner, Wray Buntine
      Description: Improving LDA topic models for microblogs.



    Jointly Leveraging Intent and Interaction Signals to Predict User Satisfaction with Slate Recommendations
    Rishabh Mehrotra, Mounia Lalmas, Doug Kenney, Thomas Lim-Meng, Golli Hashemian

    Deriving User- and Content-specific Rewards for Contextual Bandits
    Paolo Dragone, Rishabh Mehrotra, Mounia Lalmas

    The Music Streaming Sessions Dataset
    Brian Brost, Rishabh Mehrotra, Tristan Jehan

    Towards Task Understanding in Visual Settings [pdf]
    Sebastin Santy, Wazeer Zulfikar, Rishabh Mehrotra, Emine Yilmaz
    AAAI 2019 (student abstact)


    Towards a Fair Marketplace: Counterfactual Evaluation of the trade-off between Relevance, Fairness & Satisfaction in Recommendation Systems [pdf]
    Rishabh Mehrotra, James McInerney, Hugues Bouchard, Mounia Lalmas and Fernando Diaz

    Neural Attention Reader for Video Comprehension [pdf]
    Ashish Gupta, Rishabh Mehrotra and Manish Gupta
    KDD 2018, Deep Learning Day

    Explore, Exploit, and Explain: Personalizing Explainable Recommendations with Bandits
    James McInerney, Benjamin Lacker, Samantha Hansen, Karl Higley, Hugues Bouchard, Alois Gruson, Rishabh Mehrotra

    LearnIR: WSDM 2018 Workshop on Learning from User Interactions [pdf]
    Rishabh Mehrotra, Ahmed Hassan Awadallah and Emine Yilmaz


    Task Embeddings: Learning Query Embeddings using Task Context [pdf]
    Rishabh Mehrotra, Emine Yilmaz

    Deep Sequential Models for Task Satisfaction Prediction [pdf]
    Rishabh Mehrotra, Ahmed Hassan Awadallah, Milad Shokouhi, Emine Yilmaz, Imed Zitouni, Ahmed El Kholy and Madian Khabsa

    Predictive Power of Online and Offline Behavior Sequences: Evidence from a Micro-finance Context
    Rishabh Mehrotra, Prasanta Bhattacharya, Tianhui Tan and Tuan Phan

    Hey Cortana! Exploring the use cases of a Desktop based Digital Assistant [pdf]
    Rishabh Mehrotra, Ahmed Hassan Awadallah, Ahmed El Kholy and Imed Zitouni

    Characterizing and Predicting Supply-side Engagement on Crowd-contributed Video Sharing Platforms [pdf]
    Rishabh Mehrotra and Prasanta Bhattacharya

    Extracting & Leveraging User Interaction Sequences for Search Satisfaction Prediction [pdf]
    Rishabh Mehrotra, Imed Zitouni, Ahmed Hassan Awadallah and Ahmed El Kholy

    Extracting Hierarchies of Search Tasks & Subtasks via a Bayesian Nonparametric Approach [pdf]
    Rishabh Mehrotra and Emine Yilmaz

    Auditing Search Engines for Differential Performance Across Demographics [pdf]
    Rishabh Mehrotra, Ashton Anderson, Fernando Diaz, Amit Sharma, Hanna Wallach and Emine Yilmaz

    Identifying User Sessions in Interactions with Intelligent Assistants [pdf]
    Rishabh Mehrotra, Imed Zitouni, Ahmed Hassan Awadallah, Milad Shokouhi and Ahmed El Kholy
    WWW 2017 (Posters Track)


    Query Log Mining for Inferring User Tasks and Needs
    Rishabh Mehrotra and Emine Yilmaz
    ECML 2016 (Nectar Track)

    Uncovering Task Based Behavioral Heterogeneities in Online Search Behavior
    Rishabh Mehrotra, Prasanta Bhattacharya and Emine Yilmaz
    SIGIR 2016

    Characterizing Cross-Domain Search Behavior
    Rishabh Mehrotra, Paul N. Bennett, Susan Dumais, Jaime Teevan
    HIA @ SIGIR 2016

    Deconstructing Complex Search Tasks
    Rishabh Mehrotra, Prasanta Bhattacharya and Emine Yilmaz
    NAACL 2016

    Characterizing Users' Multi-Tasking Behavior in Web Search
    Rishabh Mehrotra, Prasanta Bhattacharya and Emine Yilmaz
    CHIIR 2016

    The Information Network: Exploiting Causal Dependencies in Online Information Seeking
    Prasanta Bhattacharya, Rishabh Mehrotra
    CHIIR 2016


    Representative & Informative Query Selection for Learning to Rank using Submodular Functionsd Needs
    Rishabh Mehrotra and Emine Yilmaz
    SIGIR 2015

    Terms, Topics & Tasks: Enhanced User Modelling for Better Personalization
    Rishabh Mehrotra and Emine Yilmaz
    ICTIR 2015

    Modeling the Evolution of User-generated Content on a Large Video Sharing Platform
    Rishabh Mehrotra, Prasanta Bhattacharya
    WWW 2015 (Web Science Track)

    Towards Hierarchies of Search Tasks & Subtasks
    Rishabh Mehrotra and Emine Yilmaz
    WWW 2015 (Posters track)

    Overview of the TREC 2015 Tasks Track
    Emine Yilmaz, Evangelos Kanoulas, Manisha Verma, Nick Craswell, Rishabh Mehrotra
    TREC 2015

    Topics, Tasks & Beyond: Learning Representations for Personalization
    Rishabh Mehrotra
    WSDM 2015 (Doctoral Consortium)

    A Tensor Based Approach for Coupling Search Tasks and Topical Interests for User Modelling
    Rishabh Mehrotra, Emine Yilmaz
    HIA @ WSDM 2015

    2014 & before

    Task-Based User Modelling for Personalization via Probabilistic Matrix Factorization
    Rishabh Mehrotra and Emine Yilmaz
    RecSys 2014

    Improving LDA Topic Models for Microblogs via Automatic Tweet Labeling and Pooling.
    Rishabh Mehrotra, Scott Sanner, Wray Buntine, Lexing Xie,
    SIGIR 2013

    Towards Learning Coupled Representations for Cross-Lingual Information Retrieval
    Rishabh Mehrotra, Dat Chu, Syed Aqueel Haider, Iaonnis Kakadiaris
    xLiTe @ NIPS 2012

    Dictionary based Sparse Representation for Domain Adaptation
    Rishabh Mehrotra, Rushabh Agrawal, Syed Aqueel Haider
    CIKM 2012

    Corporate News Classification and Valence Prediction: A Supervised Approach
    Syed Aqueel Haider, Rishabh Mehrotra
    WASSA @ ACL 2011

    Neural Self-Organization based Rectilinear Steiner Minimal Tree Generation in 3 Dimensions
    Rushabh Agrawal, Rishabh Mehrotra, AS Mandal
    ICMS 2012