Amit Kumar Jaiswal, Researcher at UCL

Amit Kumar Jaiswal

amitkumarj441 [at] gmail [dot] com

Github

Google Scholar

DBLP

Earlier I was a Postdoc at University College London (UCL) developing statistical machine learning and deep temporal vision models for risk stratification of Prostate cancer progression. I was a Research Engineer at University of Leeds worked on Financial AI which spin-out as 4-XTRA Technologies Ltd. I hold a PhD in Information retrieval from University of Bedfordshire, where I was a Marie Skłodowska-Curie fellow within the QUARTZ ITN consortium. I research the quantum theory evolving behind information retrieval with a focus on user-oriented IR and investigate users cognitive aspects using Information Foraging Theory. My research spans across formal information retrieval (IR) models, in particular, the mathematical formalism(s) behind Quantum Theory; Quantum Probability, Hilbert space and their usage in the study of contextual information retrieval.

Before graduate school, I earned a Bachelor's in Computer Science & Engineering from UIET, CSJM University. I was also a part of Google Summer of Code program and the Linux Foundation as an intern with Kubernetes, CNCF and Openstack team at the Open Mainframe Project.

What's new?

[Feb 2024] I'm co-organizing an WebSci'2024 workshop on "DHOW: Diffusion of Harmful Content on Online Web Workshop".
[Dec 2023] Full paper titled "FakeClaim: A Multiple Platform-driven Dataset for Identification of Fake News on 2023 Israel-Hamas War" has been accepted in ECIR 2024.
[Nov 2022] I am a visiting researcher in the Department of Radiology, University of Cambridge under Prof. Tristan Barrett.
[Sept 2022] I am serving as a program committee of the SyntheticData4ML workshop at the Conference Neural Information Processing Systems (NeurIPS) 2022.
[Sept 2022] I am serving as a program committee of the Machine Learning and the Physical Sciences workshop at the Conference Neural Information Processing Systems (NeurIPS) 2022.
[Aug 2022] Poster titled "Baseline Prediction of Prostate Cancer Progression on Active Surveillance using Clinical Parameters" has been accepted in the Early Detection of Cancer Conference 2022.
[Mar 2022] Successfully defended my PhD, here is a preprint of my thesis.
[Aug 2021] I served as a program committee of Hate Speech and Offensive Content Identification in English and Indo-Aryan Languages (HASOC) 2021 workshop at The 13th meeting of Forum for Information Retrieval Evaluation (FIRE) 2021.
[June 2021] Full paper titled "Semantic Hilbert Space for Interactive Image Retrieval" has been accepted in ACM ICTIR 2021.
[Sept 2020] I served as a program committee of the third Machine Learning and the Physical Sciences workshop at the Conference Neural Information Processing Systems (NeurIPS) 2020.
[July 2020] Full paper titled "Reinforcement Learning-driven Information Seeking: A Quantum Probabilistic Approach" has been accepted in BIRDS, SIGIR 2020.
[Dec 2019] Full paper titled "Utilising Information Foraging Theory for User Interaction with Image Query Auto-Completion" has been accepted in ECIR 2020.
[Nov 2019] I served as a reviewer of European Conference on Information Retrieval (ECIR) 2020.
[Oct 2019] Short paper titled "Information Foraging for Enhancing Implicit Feedback in Content-based Image Recommendation" has been accepted in FIRE 2019.
[Sept 2019] I served as a program committee of the 2nd Machine Learning and the Physical Sciences workshop at the Conference Neural Information Processing Systems (NeurIPS) 2019.
[Sept 2019] Received DAAD Grant to attend "Autumn School of Information Retrieval and Information Foraging 2019" at Dagstuhl [Slides].
[Aug 2019] I have been selected for the Collaborative Research Hackathon organized by the Alan Turing Institute at University of Bristol [Published Report].
[June 2019] Paper titled Effects of Foraging in Personalized Content-based Image Recommendation accepted by EARS'19, SIGIR 2019.
[Apr 2019] Visiting researcher in Department of Information Engineering, University of Padova, Italy under Prof. Massimo Melucci.
[Feb 2019] I served as a Web Chair of CHILites (SIGCHI) 2019
[Dec 2018] Presented my first year PhD progress in QUARTZ MTRM at University of Padova, Italy.
[Nov 2018] I have been selected for fee waiver scholarship, attending Search Solutions 2018 [Blogpost].
[Nov 2018] I have presented my research progress at Signal Media Research. Link to my talk.
[Oct 2018] I have been selected for full grant to attend H2O AI Conference happening in London!
[Oct 2018] I have been selected to attend Google London Meetup.
[Sept 2018] I gave a talk at Bradenburg University of Technology, Cottbus, Germany in QUARTZ Autumn School.
[Sept 2018] I gave a talk in FDIA Workshop at ICTIR 2018 on my recent paper titled "Investigating Interactive Information Retrieval via Information Foraging Theory".
[Aug 2018] I gave a talk at University of Mannheim, Germany on our recent paper.
[June 2018] I gave a talk at the Putteridge Burry at University of Bedfordshire on my progress toward Ph.D. project.
[Apr 2018] I joined University of Bedfordshire as a postgrad student. I am a part of IRAC and QUARTZ group where I work on applying quantum theory to interactive search.

Publications

FakeClaim: A Multiple Platform-Driven Dataset for Identification of Fake News on 2023 Israel-Hamas War

Gautam Kishore Shahi, Amit Kumar Jaiswal, Thomas Mandl

In Proceedings of the 46th European Conference on Information Retrieval (ECIR) (IR4Good Track 2024)

We contribute the first publicly available dataset of factual claims from different platforms and fake YouTube videos on the 2023 Israel-Hamas war for automatic fake YouTube video classification. The FakeClaim data is collected from 60 fact-checking organizations in 30 languages and enriched with metadata from the fact-checking organizations curated by trained journalists specialized in fact-checking. Further, we classify fake videos within the subset of YouTube videos using textual information and user comments. ...

[ Paper ] [ Code ] [ Poster ] [BIBTEX ]

Lightweight Adaptation of Neural Language Models via Subspace Embedding

Amit Kumar Jaiswal, Haiming Liu

In Proceedings of the 32nd ACM International Conference on Information and Knowledge Management (CIKM 2023)

Traditional neural word embeddings are usually dependent on a richer diversity of vocabulary. However, the language models recline to cover major vocabularies via the word embedding parameters, in particular, for multilingual language models that generally cover a significant part of their overall learning parameters. In this work, we present a new compact embedding structure to reduce the memory footprint of the pre-trained language models with a sacrifice of up to 4\% absolute accuracy. The embeddings vectors reconstruction follows a set of subspace embeddings and an assignment procedure via the contextual relationship among tokens from pre-trained language models. ...

[ Paper ] [ Code ] [ Poster ] [BIBTEX ]

Towards Subject Agnostic Affective Emotion Recognition

Amit Kumar Jaiswal, Haiming Liu, Prayag Tiwari

In Proceedings of the 2nd MUWS Workshop at the 32nd ACM International Conference on Information and Knowledge Management (CIKM 2023)

This paper focuses on affective emotion recognition, aiming to perform in the subject-agnostic paradigm based on EEG signals. However, EEG signals manifest subject instability in subject-agnostic affective Brain-computer interfaces (aBCIs), which led to the problem of distributional shift. Furthermore, this problem is alleviated by approaches such as domain generalisation and domain adaptation. Typically, methods based on domain adaptation confer comparatively better results than the domain generalisation methods but demand more computational resources given new subjects. We propose a novel framework, meta-learning based augmented domain adaptation for subject-agnostic aBCIs. ...

[ Paper ] [ Slides ] [BIBTEX ]

A Model-Agnostic Framework for Recommendation via Interest-aware Item Embeddings

Amit Kumar Jaiswal, Yu Xiong

In Proceedings of the 17th ACM Conference on Recommender Systems (RecSys 2023) Late-Breaking Results

Item representation holds significant importance in recommendation systems, which encompasses domains such as news, retail, and videos. Retrieval and ranking models utilise item representation to capture the user-item relationship based on user behaviours. While existing representation learning methods primarily focus on optimising item-based mechanisms, such as attention and sequential modelling. However, these methods lack a modelling mechanism to directly reflect user interests within the learned item representations. Consequently, these methods may be less effective in capturing user interests indirectly. To address this challenge, we propose a novel Interest-aware Capsule network (IaCN) recommendation model, a model-agnostic framework that directly learns interest-oriented item representations... ...