69 citations · 111 across the 27 of their papers we have counts for
30 papers
Shared Latent Structures Enable Unified Backdoor Detection and Mitigation in LLMs
Omar Mahmoud, Aly M. Kassem, Thommen George Karimpanal +4
Backdoor attacks in large language models (LLMs) are often treated as isolated trigger-response failures, motivating defenses tailored to specific triggers or behaviors. We show th…
Leveraging Human Feedback for Semantically-Relevant Skill Discovery
Maxence Hussonnois, Thommen George Karimpanal, Santu Rana
Unsupervised skill discovery in reinforcement learning aims to intrinsically motivate agents to discover diverse and useful behaviours. However, unconstrained approaches can produc…
ASPECT:Analogical Semantic Policy Execution via Language Conditioned Transfer
Ajsal Shereef Palattuparambil, Thommen George Karimpanal, Santu Rana
Reinforcement Learning (RL) agents often struggle to generalize knowledge to new tasks, even those structurally similar to ones they have mastered. Although recent approaches have…
The Unintended Trade-off of AI Alignment:Balancing Hallucination Mitigation and Safety in LLMs
Omar Mahmoud, Ali Khalil, Buddhika Laknath Semage +2
Hallucination in large language models (LLMs) has been widely studied in recent years, with progress in both detection and mitigation aimed at improving truthfulness. Yet, a critic…
MAGIK: Mapping to Analogous Goals via Imagination-enabled Knowledge Transfer
Ajsal Shereef Palattuparambil, Thommen George Karimpanal, Santu Rana
Humans excel at analogical reasoning - applying knowledge from one task to a related one with minimal relearning. In contrast, reinforcement learning (RL) agents typically require…
Beyond the Known: Decision Making with Counterfactual Reasoning Decision Transformer
Minh Hoang Nguyen, Linh Le Pham Van, Thommen George Karimpanal +2
Decision Transformers (DT) play a crucial role in modern reinforcement learning, leveraging offline datasets to achieve impressive results across various domains. However, DT requi…
Improving Multilingual Language Models by Aligning Representations through Steering
Omar Mahmoud, Buddhika Laknath Semage, Thommen George Karimpanal +1
This paper investigates how Large Language Models (LLMs) represent non-English tokens -- a question that remains underexplored despite recent progress. We propose a lightweight int…
Human-Aligned Skill Discovery: Balancing Behaviour Exploration and Alignment
Maxence Hussonnois, Thommen George Karimpanal, Santu Rana
Unsupervised skill discovery in Reinforcement Learning aims to mimic humans' ability to autonomously discover diverse behaviors. However, existing methods are often unconstrained,…
Dynamic Policy Fusion for User Alignment Without Re-Interaction
Ajsal Shereef Palattuparambil, Thommen George Karimpanal, Santu Rana
Deep reinforcement learning (RL) policies, although optimal in terms of task rewards, may not align with the personal preferences of human users. To ensure this alignment, a naive…
ECoDe: A Sample-Efficient Method for Co-Design of Robotic Agents
Kishan R. Nagiredla, Buddhika L. Semage, Arun Kumar A. +2
Co-designing autonomous robotic agents involves simultaneously optimizing the controller and physical design of the agent. Its inherent bi-level optimization formulation necessitat…
LaGR-SEQ: Language-Guided Reinforcement Learning with Sample-Efficient Querying
Thommen George Karimpanal, Laknath Buddhika Semage, Santu Rana +4
Large language models (LLMs) have recently demonstrated their impressive ability to provide context-aware responses via text. This ability could potentially be used to predict plau…
Controlled Diversity with Preference : Towards Learning a Diverse Set of Desired Skills
Maxence Hussonnois, Thommen George Karimpanal, Santu Rana
Autonomously learning diverse behaviors without an extrinsic reward signal has been a problem of interest in reinforcement learning. However, the nature of learning in such mechani…
Zero-shot Sim2Real Adaptation Across Environments
Buddhika Laknath Semage, Thommen George Karimpanal, Santu Rana +1
Simulation based learning often provides a cost-efficient recourse to reinforcement learning applications in robotics. However, simulators are generally incapable of accurately rep…
Uncertainty Aware System Identification with Universal Policies
Buddhika Laknath Semage, Thommen George Karimpanal, Santu Rana +1
Sim2real transfer is primarily concerned with transferring policies trained in simulation to potentially noisy real world environments. A common problem associated with sim2real tr…
Fast Model-based Policy Search for Universal Policy Networks
Buddhika Laknath Semage, Thommen George Karimpanal, Santu Rana +1
Adapting an agent's behaviour to new environments has been one of the primary focus areas of physics based reinforcement learning. Although recent approaches such as universal poli…
Episodic Policy Gradient Training
Hung Le, Majid Abdolshah, Thommen K. George +3
We introduce a novel training procedure for policy gradient methods wherein episodic memory is used to optimize the hyperparameters of reinforcement learning algorithms on-the-fly.…
Model-Based Episodic Memory Induces Dynamic Hybrid Controls
Hung Le, Thommen Karimpanal George, Majid Abdolshah +2
Episodic control enables sample efficiency in reinforcement learning by recalling past experiences from an episodic memory. We propose a new model-based episodic memory of trajecto…
Balanced Q-learning: Combining the Influence of Optimistic and Pessimistic Targets
Thommen George Karimpanal, Hung Le, Majid Abdolshah +4
The optimistic nature of the Q-learning target leads to an overestimation bias, which is an inherent problem associated with standard learning. Such a bias fails to account for…
Plug and Play, Model-Based Reinforcement Learning
Majid Abdolshah, Hung Le, Thommen Karimpanal George +3
Sample-efficient generalisation of reinforcement learning approaches have always been a challenge, especially, for complex scenes with many components. In this work, we introduce P…
A New Representation of Successor Features for Transfer across Dissimilar Environments
Majid Abdolshah, Hung Le, Thommen Karimpanal George +3
Transfer in reinforcement learning is usually achieved through generalisation across tasks. Whilst many studies have investigated transferring knowledge when the reward function ch…
Intuitive Physics Guided Exploration for Sample Efficient Sim2real Transfer
Buddhika Laknath Semage, Thommen George Karimpanal, Santu Rana +1
Physics-based reinforcement learning tasks can benefit from simplified physics simulators as they potentially allow near-optimal policies to be learned in simulation. However, such…
Neuro-evolutionary Frameworks for Generalized Learning Agents
Thommen George Karimpanal
The recent successes of deep learning and deep reinforcement learning have firmly established their statuses as state-of-the-art artificial learning techniques. However, longstandi…
Learning Transferable Domain Priors for Safe Exploration in Reinforcement Learning
Thommen George Karimpanal, Santu Rana, Sunil Gupta +2
Prior access to domain knowledge could significantly improve the performance of a reinforcement learning agent. In particular, it could help agents avoid potentially catastrophic e…
Self-Organizing Maps for Storage and Transfer of Knowledge in Reinforcement Learning
Thommen George Karimpanal, Roland Bouffanais
The idea of reusing or transferring information from previously learned tasks (source tasks) for the learning of new tasks (target tasks) has the potential to significantly improve…