Hi, I'm Pranav.

I'm a doctoral researcher at the University of Oxford advised by Prof. Ben Seymour (NDCN, IBME) and co-advised by Prof. Ioannis Havoutis (ORI). My background is in engineering and I'm a Neuro-AI researcher working on fundamental problems with applications to safe AI and AI for healthcare. I previously interned at Nvidia where I worked on scalable distributed deep learning.

I'm introverted. I'm a cheerful pessimist. I try to keep thinking about conflicting ideas for longer periods and I find paradoxes deeply interesting. Much of my work is very interdisciplinary and I'm quite comfortable being an outsider in a new field of study.

You can read my sporadically updated blog here :)

News

I'll be presenting a Spotlight presentation and a poster on our work on Quantitative Motor Testing at the Podium Institute Conference.
My work on Neural Associative Skill Memories has been accepted at 5th International Workshop on Active Inference (IWAI 2024) as a Spotlight presentation.
My DPhil's first preprint "Balancing safety and efficiency in human decision making" is now in eLife.
I will be giving an invited talk at 1st Oxford Health BRC Pain Conference (March 2024) on ’Virtual reality and Pain: From theory to practice’

Research vision

Key motivations in Neuro-AI research are two-fold, using advancements from AI to generate useful hypotheses for neuroscience and using insights from neuroscience to develop better algorithms for AI. With these motivations in mind, my research spans the following three directions:

Safe Natural Intelligence
- How do animals keep themselves safe? (e.g. safe exploration, self-preservation)
- How to reverse-engineer our pain system?
Safe Artificial Intelligence
- How can we make intelligent machines that act safely in human-like ways?
- How can we make AI trustworthy, interpretable, and aligned with human values in human-AI interactions?
AI and computational neuroscience for healthcare
- New technologies for better assessment and treatments to improve human health and well-being (with a recent focus on chronic pain)

Selected Publications

Safe exploration and human reinforcement learning

Optimal composition of multiple values for dopamine-mediated efficient, safe and stable learning

RLDM conference (2025) and in preparation

Pranav Mahajan, Ben Seymour

Developed a theory that enables optimal composition of multiple value functions to drive reliable behaviour and efficient off-policy adaptation to dynamically changing rewards
Reconciled contradicting hypotheses about the computation of the tail of the striatum (threat prediction error and action prediction error), providing a unifying normative account

Balancing safety and efficiency in human decision making

eLife (2024) and COSYNE conference (2022)

Pranav Mahajan, Shuangyi Tong, Sang Wan Lee, Ben Seymour

Formulated a safety-efficiency trade-off in human reinforcement learning, proposing an uncertainty-based solution.
Designed and performed a Virtual Reality experiment to test the model and performed hierarchical Bayesian modeling of human behavior.

Webpage

Paper

BibTex

Computational neuroscience and neuro-robotics

Neural Associative Skill Memories for safer robotics and modelling human sensorimotor repertoires

5th International Workshop on Active Inference (IWAI 2024) - Spotlight and in preparation

Pranav Mahajan, Mufeng Tang, T. Ed Li, Ioannis Havoutis, Ben Seymour

Developed embodied generative models learned from demonstrations (in PyTorch) for safer robots (Franka Emika Panda arm), implementing fault detection and correction while explaining human sensorimotor experiment findings

Homeostasis After Injury: How Intertwined Inference and Control Underpin Post-Injury Pain and Behaviour

Under review at CCN (2025)

Pranav Mahajan, Peter Dayan, Ben Seymour

Theorised homeostasis after injury as planning under partial observability, explaining paradoxes and providing a novel perspective on existing literature.

Paper

BibTex

Enhanced behavioural and neural sensitivity to punishments in chronic pain and fatigue

Brain (2024)

Flavia Mancini, Pranav Mahajan, Anna á V Guttesen, Jakub Onysk, Ingrid Scholtes, Nicholas Shenker, Michael Lee, Ben Seymour

Analyzed chronic pain patient high-dimensional brain data using graph-theoretic measures to predict pain and fatigue levels from insular cortical networks.

Paper

BibTex

Doing what’s not wanted: Conflict in incentives and misallocation of behavioural control can lead to drug-seeking despite adverse outcomes

Addiction Neuroscience (2023)

Pranav Mahajan, Veeky Baths, Boris Gutkin

Algorithmic modelling of addiction as drug-induced hijacking of a hierarchical reinforcement learning algorithm

Paper

BibTex

Quantifying Synchronization in a Biologically Inspired Neural Network

IEEE- International Joint Conference on Neural Networks (IJCNN 2021) and Bernstein Conference (2020)

Pranav Mahajan, Advait Rane, Swapna Sasi, Basabdatta Sen Bhattacharya

Developed a toolbox for quantifying synchronization in multi-variate time-series data.

SyncBox

Paper

BibTex

AI for healthcare

Quantitative Movement Testing (QMT)

1st Podium Institute Conference (2024) - Spotlight and in preparation

Pranav Mahajan, Amanda Wall, Eoin Kelleher, Anushka Soni, Ben Seymour

Developed a computer vision pipeline to track movements from handheld video, securing £3,000 in funding for healthy subject and patient trials

Webpage

Quantitative Cognitive Aversive Testing (QCAT)

Neuromatch Conference (2022)

Pranav Mahajan, Jakub Onysk, Suyi Zhang, Katja Wiech, Flavia Mancini, Ben Seymour

Prepared a suite of mobile-based cognitive tests for pain patients whose behaviour can be modelled using reinforcement learning algorithms

Webpage

Slides

Acoustic and Language Based Deep Learning Approaches for Alzheimer's Dementia Detection From Spontaneous Speech

Frontiers in Aging Neuroscience (2021)

Pranav Mahajan, Veeky Baths

Developed a multimodal deep learning model enriched with targeted features for early Alzheimer’s detection from speech, achieving a 6.25% improvement in accuracy.

Paper

BibTex

Archived Projects

Hi, I'm Pranav.

News

Research vision

Selected Publications

Safe exploration and human reinforcement learning

Optimal composition of multiple values for dopamine-mediated efficient, safe and stable learning

Balancing safety and efficiency in human decision making

Computational neuroscience and neuro-robotics

Neural Associative Skill Memories for safer robotics and modelling human sensorimotor repertoires

Homeostasis After Injury: How Intertwined Inference and Control Underpin Post-Injury Pain and Behaviour

Enhanced behavioural and neural sensitivity to punishments in chronic pain and fatigue

Doing what’s not wanted: Conflict in incentives and misallocation of behavioural control can lead to drug-seeking despite adverse outcomes

Quantifying Synchronization in a Biologically Inspired Neural Network

AI for healthcare

Quantitative Movement Testing (QMT)

Quantitative Cognitive Aversive Testing (QCAT)

Acoustic and Language Based Deep Learning Approaches for Alzheimer's Dementia Detection From Spontaneous Speech

Want to get in touch?

Email me at pranav[dot]mahajan[at]ndcn.ox.ac.uk