Task-specific Subnetwork Discovery in Reinforcement Learning for Autonomous Underwater Navigation

Autonomous underwater vehicles are required to perform multiple tasks adaptively and in an explainable manner under dynamic, uncertain conditions and limited sensing, challenges that classical controllers struggle to address. This demands robust, generalizable, and inherently interpretable control policies for reliable long-term monitoring. Reinforcement learning, particularly multi-task RL, overcomes these limitations by leveraging shared representations to enable efficient adaptation across tasks and environments. However, while such policies show promising results in simulation and controlled experiments, they yet remain opaque and offer limited insight into the agent's internal decision-making, creating gaps in transparency, trust, and safety that hinder real-world deployment. The internal policy structure and task-specific specialization remain poorly understood. To address these gaps, we analyze the internal structure of a pretrained multi-task reinforcement learning network in the HoloOcean simulator for underwater navigation by identifying and comparing task-specific subnetworks responsible for navigating toward different species. We find that in a contextual multi-task reinforcement learning setting with related tasks, the network uses only about 1.5% of its weights to differentiate between tasks. Of these, approximately 85% connect the context-variable nodes in the input layer to the next hidden layer, highlighting the importance of context variables in such settings. Our approach provides insights into shared and specialized network components, useful for efficient model editing, transfer learning, and continual learning for underwater monitoring through a contextual multi-task reinforcement learning method.

DINO-Explorer: Active Underwater Discovery via Ego-Motion Compensated Semantic Predictive Coding

Marine ecosystem degradation necessitates continuous, scientifically selective underwater monitoring. However, most autonomous underwater vehicles (AUVs) operate as passive data loggers, capturing exhaustive video for offline review and frequently missing transient events of high scientific value. Transitioning to active perception requires a causal, online signal that highlights significant phenomena while suppressing maneuver-induced visual changes. We propose DINO-Explorer, a novelty-aware perception framework driven by a continuous semantic surprise signal. Operating within the latent space of a frozen DINOv3 foundation model, it leverages a lightweight, action-conditioned recurrent predictor to anticipate short-horizon semantic evolution. An efference-copy-inspired module utilizes globally pooled optical flow to discount self-induced visual changes without suppressing genuine environmental novelty. We evaluate this signal on the downstream task of asynchronous event triage under variant telemetry constraints. Results demonstrate that DINO-Explorer provides a robust, bandwidth-efficient attention mechanism. At a fixed operating point, the system retains 78.8% of post-discovery human-reviewer consensus events with a 56.8% trigger confirmation rate, effectively surfacing mission-relevant phenomena. Crucially, ego-motion conditioning suppresses 45.5% of false positives relative to an uncompensated surprise signal baseline. In a replay-side Pareto ablation study, DINO-Explorer robustly dominates the validated peak F1 versus telemetry bandwidth frontier, reducing telemetry bandwidth by 48.2% at the selected operating point while maintaining a 62.2% peak F1 score, successfully concentrating data transmission around human-verified novelty events.

Contextual Multi-Task Reinforcement Learning for Autonomous Reef Monitoring

Although autonomous underwater vehicles promise the capability of marine ecosystem monitoring, their deployment is fundamentally limited by the difficulty of controlling vehicles under highly uncertain and non-stationary underwater dynamics. To address these challenges, we employ a data-driven reinforcement learning approach to compensate for unknown dynamics and task variations.Traditional single-task reinforcement learning has a tendency to overfit the training environment, thus, limit the long-term usefulness of the learnt policy. Hence, we propose to use a contextual multi-task reinforcement learning paradigm instead, allowing us to learn controllers that can be reused for various tasks, e.g., detecting oysters in one reef and detecting corals in another. We evaluate whether contextual multi-task reinforcement learning can efficiently learn robust and generalisable control policies for autonomous underwater reef monitoring. We train a single context-dependent policy that is able to solve multiple related monitoring tasks in a simulated reef environment in HoloOcean. In our experiments, we empirically evaluate the contextual policies regarding sample-efficiency, zero-shot generalisation to unseen tasks, and robustness to varying water currents. By utilising multi-task reinforcement learning, we aim to improve the training effectiveness, as well as the reusability of learnt policies to take a step towards more sustainable procedures in autonomous reef monitoring.

Differentiable Inverse Graphics for Zero-shot Scene Reconstruction and Robot Grasping

Operating effectively in novel real-world environments requires robotic systems to estimate and interact with previously unseen objects. Current state-of-the-art models address this challenge by using large amounts of training data and test-time samples to build black-box scene representations. In this work, we introduce a differentiable neuro-graphics model that combines neural foundation models with physics-based differentiable rendering to perform zero-shot scene reconstruction and robot grasping without relying on any additional 3D data or test-time samples. Our model solves a series of constrained optimization problems to estimate physically consistent scene parameters, such as meshes, lighting conditions, material properties, and 6D poses of previously unseen objects from a single RGBD image and bounding boxes. We evaluated our approach on standard model-free few-shot benchmarks and demonstrated that it outperforms existing algorithms for model-free few-shot pose estimation. Furthermore, we validated the accuracy of our scene reconstructions by applying our algorithm to a zero-shot grasping task. By enabling zero-shot, physically-consistent scene reconstruction and grasping without reliance on extensive datasets or test-time sampling, our approach offers a pathway towards more data efficient, interpretable and generalizable robot autonomy in novel environments.

Bayesian Inverse Physics for Neuro-Symbolic Robot Learning

Real-world robotic applications, from autonomous exploration to assistive technologies, require adaptive, interpretable, and data-efficient learning paradigms. While deep learning architectures and foundation models have driven significant advances in diverse robotic applications, they remain limited in their ability to operate efficiently and reliably in unknown and dynamic environments. In this position paper, we critically assess these limitations and introduce a conceptual framework for combining data-driven learning with deliberate, structured reasoning. Specifically, we propose leveraging differentiable physics for efficient world modeling, Bayesian inference for uncertainty-aware decision-making, and meta-learning for rapid adaptation to new tasks. By embedding physical symbolic reasoning within neural models, robots could generalize beyond their training data, reason about novel situations, and continuously expand their knowledge. We argue that such hybrid neuro-symbolic architectures are essential for the next generation of autonomous systems, and to this end, we provide a research roadmap to guide and accelerate their development.

Bayesian Inverse Graphics for Few-Shot Concept Learning

Humans excel at building generalizations of new concepts from just one single example. Contrary to this, current computer vision models typically require large amount of training samples to achieve a comparable accuracy. In this work we present a Bayesian model of perception that learns using only minimal data, a prototypical probabilistic program of an object. Specifically, we propose a generative inverse graphics model of primitive shapes, to infer posterior distributions over physically consistent parameters from one or several images. We show how this representation can be used for downstream tasks such as few-shot classification and pose estimation. Our model outperforms existing few-shot neural-only classification algorithms and demonstrates generalization across varying lighting conditions, backgrounds, and out-of-distribution shapes. By design, our model is uncertainty-aware and uses our new differentiable renderer for optimizing global scene parameters through gradient descent, sampling posterior distributions over object parameters with Markov Chain Monte Carlo (MCMC), and using a neural based likelihood function.