Joni Pajarinen

Latest

• Automating on-site object inspection with a quadruped robot and BIM
• Efficient Reinforcement Learning by Guiding World Models with Non-Curated Data
• Embodied Active Learning under Limited Annotation and Navigation Budget for Object Detection
• Learning Multi-Timescale Abstractions for Hierarchical Combinatorial Planning
• Modesuite: Robot learning task suite for benchmarking mobile manipulation with deformable objects
• Object-Centric Vision Token Pruning for Vision Language Models
• Offline Meta-Reinforcement Learning in Piecewise Stationary Environments
• Pareto-Conditioned Diffusion Models for Offline Multi-Objective Optimization
• Rethinking Temporal Consistency in Video Object-Centric Learning: From Prediction to Correspondence
• Smoothing Slot Attention Iterations and Recurrences
• MetaSlot: Break Through the Fixed Number of Slots in Object-Centric Learning
• Manipulate-to-Navigate: Reinforcement Learning with Visual Affordances and Manipulability Priors
• Probabilistic Subgoal Representations for Hierarchical Reinforcement learning
• Extracting Visual Plans from Unlabeled Videos via Symbolic Guidance
• Vector-Quantized Vision Foundation Models for Object-Centric Learning
• Discrete Codebook World Models for Continuous Control
• Multi-Scale Fusion for Object Representation
• AgentMixer: Multi-Agent Correlated Policy Factorization
• Entropy Regularized Task Representation Learning for Offline Meta-Reinforcement Learning
• End-Effector Cartesian Velocity Control for Redundant Loader Cranes Using Reinforcement learning
• On the benefit of optimal transport for curriculum reinforcement learning
• Backpropagation Through Agents
• Optimistic Multi-Agent Policy Gradient
• RP1M: A Large-Scale Motion Dataset for Piano Playing with Bi-Manual Dexterous Robot Hands
• Continuous Monte Carlo Graph Search
• Function-space Prameterization of Neural Networks for Sequential Learning
• Probabilistic Subgoal Representations for Hierarchical Reinforcement learning
• Residual Learning and Context Encoding for Adaptive Offline-to-Online Reinforcement Learning
• Hybrid search for efficient planning with completeness guarantees
• Hierarchical Imitation Learning with Vector Quantized Models
• Simplified Temporal Consistency Reinforcement Learning
• Nonlinear Model Learning for Compensation and Feedforward Control of Real-World Hydraulic Actuators Using Gaussian Processes
• POMDP planning under object composition uncertainty: Application to robotic manipulation
• Curriculum reinforcement learning via constrained optimal transport
• Partially Observable Markov Decision Processes in Robotics: A Survey
• Boosted Curriculum Reinforcement Learning
• GPU-accelerated policy optimization via batch automatic differentiation of gaussian processes for real-world control
• Reinforcement Learning (RL) Reading Club
• A Probabilistic Interpretation of Self-Paced Learning with Applications to Reinforcement Learning
• Convex Regularization in Monte-Carlo Tree Search
• Self-Paced Deep Reinforcement Learning
• Generalized Mean Estimation in Monte-Carlo Tree Search
• Multi-agent active information gathering in discrete and continuous-state decentralized POMDPs by policy graph improvement
• Probabilistic approach to physical object disentangling
• Compatible natural gradient policy search
• Projections for Approximate Policy Iteration Algorithms
• Learning Intention Aware Online Adaptation of Movement Primitives
• An Algorithmic Perspective on Imitation Learning
• Hybrid Control Trajectory Optimization under Uncertainty
• Robotic manipulation of multiple objects as a POMDP