Siddharth Bhatia

I am co-founder of Puch AI, building AI for Billions.

Before founding Puch AI, I built TurboML a platform for building AI that is Continual, Adaptive, and Personalized. I completed my PhD at National University of Singapore (NUS) advised by Bryan Hooi and bachelors at BITS Pilani. My research was supported by the Presidents Graduate Fellowship and I was recognized as a Young Researcher by the ACM Heidelberg Laureate Forum.

During my PhD, I spent wonderful summers interning at Google Research and Amazon Web Services. I also co-organized the Outlier Detection and Description (ODD) workshop at KDD which brings together academic and industry researchers, and practitioners to discuss and reflect on outlier mining challenges.

Ph.D. Thesis

Streaming Anomaly Detection
[PDF] [Github] [Slides]

Select Publications

  1. MIDAS: Microcluster-Based Detector of Anomalies in Edge Streams [500+ stars on GitHub]
    Siddharth Bhatia, Bryan Hooi, Minji Yoon, Kijung Shin, Christos Faloutsos
    AAAI Conference on Artificial Intelligence (AAAI), 2020
    [Paper] [Code] [Overview Video]

  2. MStream: Fast Anomaly Detection in Multi-Aspect Streams [Best Paper Candidate]
    Siddharth Bhatia, Arjit Jain, Pan Li, Ritesh Kumar, Bryan Hooi
    The Web Conference (WWW), 2021
    [Paper] [Code]

  3. MemStream: Memory-Based Streaming Anomaly Detection
    Siddharth Bhatia, Arjit Jain, Shivin Srivastava, Kenji Kawaguchi, Bryan Hooi
    The Web Conference (WWW), 2022
    [Paper] [Code]

  4. Sketch-Based Anomaly Detection in Streaming Graphs
    Siddharth Bhatia, Mohit Wadhwa, Kenji Kawaguchi, Neil Shah, Philip S. Yu, Bryan Hooi
    SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), 2023
    [Paper] [Code]

  5. ExGAN: Adversarial Generation of Extreme Samples
    Siddharth Bhatia*, Arjit Jain*, Bryan Hooi [ * denotes equal contribution]
    AAAI Conference on Artificial Intelligence (AAAI), 2021
    [Paper] [Code]

  6. SSMF: Shifting Seasonal Matrix Factorization
    Koki Kawabata*, Siddharth Bhatia*, Rui Liu, Mohit Wadhwa, Bryan Hooi [ * denotes equal contribution]
    Conference on Neural Information Processing Systems (NeurIPS), 2021
    [Paper] [Code]

News

Nov, 2022 Defended my Ph.D!
Aug, 2022 Submitted my Ph.D. Thesis. Honoured to receive the Research Achievement Award at NUS.
Jan, 2022 MemStream was accepted in The Web Conference (WWW) 2022. Code on Github.
Oct, 2021 MIDAS extended journal version accepted in TKDD.
Oct, 2021 Monlad was accepted in WSDM 2022.
Sep, 2021 Two papers accepted at NeurIPS 2021.
Aug, 2021 Co-organising the Outlier Detection and Description (ODD) workshop at KDD 2021.
Aug, 2021 Honoured to receive the Research Achievement Award at NUS.
May, 2021 Interning with Google Research during the summer.
May, 2021 Invited to give a seminar at MIT.
Mar, 2021 Invited to give seminars at Alan Turing Institute and NYU.
Feb, 2021 KDnuggets covered ExGAN: Adversarial Generation of Extreme Samples.
Jan, 2021 MStream was accepted in The Web Conference (WWW) 2021. Code on Github.
Dec, 2020 ExGAN was accepted in AAAI 2021. Code on Github.
Dec, 2020 AugSplicing was accepted in AAAI 2021. Code on Github.
Oct, 2020 Invited to speak at the Data+AI Summit hosted by Databricks.
Oct, 2020 AIhub covered ExGAN: Adversarial Generation of Extreme Samples.
Aug, 2020 Serving as the Publications Chair for SOCC 2020.
Aug, 2020 Honoured to receive the Research Achievement Award at NUS.
Aug, 2020 Invited to speak at the DataScience SG meetup and RSA Conference.
Jul, 2020 MIDAS wins popular choice award at Microsoft Azure Hackathon.
Jun, 2020 Invited to speak at the Knowledge Graphs Meetup.
Jun, 2020 Interning with AI Labs at Amazon Web Services during the summer.
Jun, 2020 Invited to speak at the Data Science Congress.
May, 2020 Interviewed by AIhub.
Apr, 2020 Zak Jost from Amazon invited me to present my research. Recording can be viewed on Youtube.
Apr, 2020 MIDAS was featured as a top story on KDnuggets.
Mar, 2020 MIDAS was covered by Towards Data Science, Towards AI, and ACM TechNews.
Nov, 2019 MIDAS was accepted in AAAI 2020. It detects microcluster anomalies in dynamic graphs. Code on Github.

Contact Information

E-mail [email protected]