177 citations · 280 across the 7 of their papers we have counts for
8 papers
A Safety and Security Framework for Real-World Agentic Systems
Shaona Ghosh, Barnaby Simkin, Kyriacos Shiarlis +9
This paper introduces a dynamic and actionable framework for securing agentic AI systems in enterprise deployment. We contend that safety and security are not merely fixed attribut…
Gandalf the Red: Adaptive Security for LLMs
Niklas Pfister, Václav Volhejn, Manuel Knott +23
Current evaluations of defenses against prompt attacks in large language model (LLM) applications often overlook two critical factors: the dynamic nature of adversarial behavior an…
Hierarchical Imitation Learning for Stochastic Environments
Maximilian Igl, Punit Shah, Paul Mougin +5
Many applications of imitation learning require the agent to generate the full distribution of behaviour observed in the training data. For example, to evaluate the safety of auton…
Symphony: Learning Realistic and Diverse Agents for Autonomous Driving Simulation
Maximilian Igl, Daewoo Kim, Alex Kuefler +7
Simulation is a crucial tool for accelerating the development of autonomous vehicles. Making simulation realistic requires models of the human road users who interact with such car…
VariBAD: A Very Good Method for Bayes-Adaptive Deep RL via Meta-Learning
Luisa Zintgraf, Kyriacos Shiarlis, Maximilian Igl +4
Trading off exploration and exploitation in an unknown environment is key to maximising expected return during learning. A Bayes-optimal policy, which does so optimally, conditions…
Learning from Demonstration in the Wild
Feryal Behbahani, Kyriacos Shiarlis, Xi Chen +8
Learning from demonstration (LfD) is useful in settings where hand-coding behaviour or a reward function is impractical. It has succeeded in a wide range of problems but typically…
Fast Context Adaptation via Meta-Learning
Luisa M Zintgraf, Kyriacos Shiarlis, Vitaly Kurin +2
We propose CAVIA for meta-learning, a simple extension to MAML that is less prone to meta-overfitting, easier to parallelise, and more interpretable. CAVIA partitions the model par…
TACO: Learning Task Decomposition via Temporal Alignment for Control
Kyriacos Shiarlis, Markus Wulfmeier, Sasha Salter +2
Many advanced Learning from Demonstration (LfD) methods consider the decomposition of complex, real-world tasks into simpler sub-tasks. By reusing the corresponding sub-policies wi…