Rudra P.K. Poudel

Rudra P.K. Poudel
Associate Principal Research Scientist
Cambridge Research Laboratory
Toshiba Europe, Cambridge, UK

My research interests lie at the intersection of artificial intelligence and cognitive science. I am particularly interested to discover how we learn and how we see. Can we enable machines to do the same things? To answer these questions I am working in machine learning, computer vision and robotics research. My current research interests include multimodal representation learning, world models, reinforcement learning, continual learning, embodied AI, and AI for social good.

Links : CV, Google scholar, Github, LinkedIn, Twitter

News

Invited speaker at ECCV Workshop 2026 on Safe World Models for Trustworthy Embodied AI
Invited talk at University of Oxford, 8 Nov 2024 on Multimodal World Models for Embodied Agents
Invited talk at AI Accelerator Institute, London, 7 Nov 2024 on Multimodal World Models for Embodied Agents
Organizing a BMVA Symposium on Robotics Foundation & World Models, 30 October 2024
Invited talk at Imperial College London, 31 May 2024 on World Models for Embodied Agents
Senior program committee member of the AAAI Conference on Artificial Intelligence, AAAI-22

Highlights

World Models, BMVA Symposium Invited Talk, 2024

AI for Developing Countries, NEA Keynote, 2023

Selected Publications

eg. Value Conditioned Policy Fine Tuning for Test Time Domain Adaptation

Value Conditioned Policy Fine Tuning for Test Time Domain Adaptation
International Conference on Machine Learning (ICML) Workshop, 2025
Harit Pandya, Ignas Budvytis, Rudra P.K. Poudel, Stephan Liwicki
[ICML Workshop on PUT]

Surgical Vision World Model
arXiv, 2025
Saurabh Koju, Saurav Bastola, Prashant Shrestha, Sanskar Amgain, Yash Raj Shrestha, Rudra P.K. Poudel, Binod Bhattarai
[arXiv]

eg. ReCoRe: Regularized Contrastive Representation Learning of World Model

ReCoRe: Regularized Contrastive Representation Learning of World Model
IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024
Rudra P.K. Poudel, Harit Pandya, Stephan Liwicki, Roberto Cipolla
[Project Page] [Press Release] [CVPR] [arXiv] [Presentation Video] [Code]

Presentation

eg. Generalized Autonomous Optimization for Quantum Transmitters with Deep Reinforcement Learning

Generalized Autonomous Optimization for Quantum Transmitters with Deep Reinforcement Learning
Photonics West: Quantum Computing, Communication, and Simulation, 2024
Yuen San Lo, Robert Woodward, Taofiq Paraiso, Rudra P.K. Poudel, Andrew Shields
[Photonics West] [pdf]

eg. LanGWM: Language Grounded World Model

LanGWM: Language Grounded World Model
arXiv, 2023
Rudra P.K. Poudel, Harit Pandya, Chao Zhang, Roberto Cipolla
[arXiv]

eg. Contrastive Unsupervised Learning of World Model with Invariant Causal Features

Contrastive Unsupervised Learning of World Model with Invariant Causal Features
Neural Information Processing Systems (NeurIPS) Workshop, 2022
Rudra P.K. Poudel, Harit Pandya, Roberto Cipolla
[arXiv] [NeurIPS Workshop on CML4Impact]

Training Binary Neural Networks the Easy Way
British Machine Vision Conference (BMVC), 2022
Alasdair Paren, Rudra P.K. Poudel
[BMVC] [BMVC Supplementary]

CoMBiNED: Multi-Constrained Model Based Planning for Navigation in Dynamic Environments
IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2022
Harit Pandya, Rudra P.K. Poudel, Stephan Liwicki
[ICML Workshop]

Faking Interpolation Until You Make It
Transactions of Machine Learning Research (TMLR), 2022
Alasdair Paren, Rudra P.K. Poudel, M Pawan Kumar
[TMLR]

A Stochastic Bundle Method for Interpolation
The Journal of Machine Learning Research (JMLR), 2022
Alasdair Paren, Leonard Berrada, Rudra P.K. Poudel, M Pawan Kumar
[JMLR]

Embodied Visual Navigation with Automatic Curriculum Learning in Real Environments
IEEE Robotics and Automation Letters (RA-L), 2020
Steven D Morad, Roberto Mecca, Rudra P.K. Poudel, Stephan Liwicki, Roberto Cipolla
[pdf] [IEEE RA-L] [arXiv]

Fast-SCNN: Fast Semantic Segmentation Network
British Machine Vision Conference (BMVC), 2019 [Oral]
Rudra P.K. Poudel, Stephan Liwicki, Roberto Cipolla
[pdf] [Slides] [arXiv]

ContextNet: Exploring Context and Detail for Semantic Segmentation in Real-time
British Machine Vision Conference (BMVC), 2018 [Oral]
Rudra P.K. Poudel, Ujwal Bonde, Stephan Liwicki, Christopher Zach
[pdf] [Slides] [arXiv] [Code]

eg. Predicting Brain Age with Deep Learning from Raw Imaging Data Results in a Reliable and Heritable Biomarker

Predicting Brain Age with Deep Learning from Raw Imaging Data Results in a Reliable and Heritable Biomarker
NeuroImage, 2017
James H Cole, Rudra P.K. Poudel, Dimosthenis Tsagkrasoulis, Matthan WA Caan, Claire Steves, Tim D Spector, Giovanni Montana
[pdf] [NeuroImage] [arXiv] [MIT Tech Review]

eg. Recurrent Fully Convolutional Neural Networks for Multi-slice MRI Cardiac Segmentation

Recurrent Fully Convolutional Neural Networks for Multi-slice MRI Cardiac Segmentation
International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) Workshop, 2016
Rudra P.K. Poudel, Pablo Lamata, Giovanni Montana
[pdf] [arXiv]

eg. A Unified Framework for 3D Hand Tracking

A Unified Framework for 3D Hand Tracking
International Symposium on Visual Computing (ISVC), 2013 [Oral]
Rudra P.K. Poudel, Jose AS Fonseca, Jian J Zhang, Hammadi Nait-Charif
[pdf] [Demo Video]

Demo

View All Publications >

Teaching

Current:

No Classes

Past:

Object Oriented Programming with C++
Management Information System
Operations Management

Mentoring:

PhD,

Alasdair Paren, PhD in University of Oxford, jointly with M Pawan Kumar, 2018-2022
Anthony Hu, PhD in University of Cambridge, jointly with Roberto Cipolla, 2018-2022

MPhil,

Uras Asil, University of Cambridge, 2026
Nanze Chen, University of Cambridge, 2024

Intern,

Ruiqi Zhu, King´s College London, 2024
Alasdair Paren, University of Oxford, 2021
Steven Morad, University of Arizona, 2020