Joni Pajarinen

Yi Zhao, Rinu Boney, Alexander Ilin et al.

Offline reinforcement learning, by learning from a fixed dataset, makes it possible to learn agent behaviors without interacting with the environment. However, depending on the quality of the offline dataset, such pre-trained agents may have limited performance and would further need to be fine-tuned online by interacting with the environment. During online fine-tuning, the performance of the pre-trained agent may collapse quickly due to the sudden distribution shift from offline to online data. While constraints enforced by offline RL methods such as a behaviour cloning loss prevent this to an extent, these constraints also significantly slow down online fine-tuning by forcing the agent to stay close to the behavior policy. We propose to adaptively weigh the behavior cloning loss during online fine-tuning based on the agent's performance and training stability. Moreover, we use a randomized ensemble of Q functions to further increase the sample efficiency of online fine-tuning by performing a large number of learning updates. Experiments show that the proposed method yields state-of-the-art offline-to-online reinforcement learning performance on the popular D4RL benchmark. Code is available: \url{https://github.com/zhaoyi11/adaptive_bc}.

Eric Chen, Zhang-Wei Hong, Joni Pajarinen et al.

State-of-the-art reinforcement learning (RL) algorithms typically use random sampling (e.g., $ε$-greedy) for exploration, but this method fails on hard exploration tasks like Montezuma's Revenge. To address the challenge of exploration, prior works incentivize exploration by rewarding the agent when it visits novel states. Such intrinsic rewards (also called exploration bonus or curiosity) often lead to excellent performance on hard exploration tasks. However, on easy exploration tasks, the agent gets distracted by intrinsic rewards and performs unnecessary exploration even when sufficient task (also called extrinsic) reward is available. Consequently, such an overly curious agent performs worse than an agent trained with only task reward. Such inconsistency in performance across tasks prevents the widespread use of intrinsic rewards with RL algorithms. We propose a principled constrained optimization procedure called Extrinsic-Intrinsic Policy Optimization (EIPO) that automatically tunes the importance of the intrinsic reward: it suppresses the intrinsic reward when exploration is unnecessary and increases it when exploration is required. The results is superior exploration that does not require manual tuning in balancing the intrinsic reward against the task reward. Consistent performance gains across sixty-one ATARI games validate our claim. The code is available at https://github.com/Improbable-AI/eipo.

Zhang-Wei Hong, Aviral Kumar, Sathwik Karnik et al.

Offline policy learning is aimed at learning decision-making policies using existing datasets of trajectories without collecting additional data. The primary motivation for using reinforcement learning (RL) instead of supervised learning techniques such as behavior cloning is to find a policy that achieves a higher average return than the trajectories constituting the dataset. However, we empirically find that when a dataset is dominated by suboptimal trajectories, state-of-the-art offline RL algorithms do not substantially improve over the average return of trajectories in the dataset. We argue this is due to an assumption made by current offline RL algorithms of staying close to the trajectories in the dataset. If the dataset primarily consists of sub-optimal trajectories, this assumption forces the policy to mimic the suboptimal actions. We overcome this issue by proposing a sampling strategy that enables the policy to only be constrained to ``good data" rather than all actions in the dataset (i.e., uniform sampling). We present a realization of the sampling strategy and an algorithm that can be used as a plug-and-play module in standard offline RL algorithms. Our evaluation demonstrates significant performance gains in 72 imbalanced datasets, D4RL dataset, and across three different offline RL algorithms. Code is available at https://github.com/Improbable-AI/dw-offline-rl.

Mikko Lauri, David Hsu, Joni Pajarinen

Noisy sensing, imperfect control, and environment changes are defining characteristics of many real-world robot tasks. The partially observable Markov decision process (POMDP) provides a principled mathematical framework for modeling and solving robot decision and control tasks under uncertainty. Over the last decade, it has seen many successful applications, spanning localization and navigation, search and tracking, autonomous driving, multi-robot systems, manipulation, and human-robot interaction. This survey aims to bridge the gap between the development of POMDP models and algorithms at one end and application to diverse robot decision tasks at the other. It analyzes the characteristics of these tasks and connects them with the mathematical and algorithmic properties of the POMDP framework for effective modeling and solution. For practitioners, the survey provides some of the key task characteristics in deciding when and how to apply POMDPs to robot tasks successfully. For POMDP algorithm designers, the survey provides new insights into the unique challenges of applying POMDPs to robot systems and points to promising new directions for further research.

Rongzhen Zhao, Vivienne Wang, Juho Kannala et al.

Object-Centric Learning (OCL) represents dense image or video pixels as sparse object features. Representative methods utilize discrete representation composed of Variational Autoencoder (VAE) template features to suppress pixel-level information redundancy and guide object-level feature aggregation. The most recent advancement, Grouped Discrete Representation (GDR), further decomposes these template features into attributes. However, its naive channel grouping as decomposition may erroneously group channels belonging to different attributes together and discretize them as sub-optimal template attributes, which losses information and harms expressivity. We propose Organized GDR (OGDR) to organize channels belonging to the same attributes together for correct decomposition from features into attributes. In unsupervised segmentation experiments, OGDR is fully superior to GDR in augmentating classical transformer-based OCL methods; it even improves state-of-the-art diffusion-based ones. Codebook PCA and representation similarity analyses show that compared with GDR, our OGDR eliminates redundancy and preserves information better for guiding object representation learning. The source code is available in the supplementary material.

Pascal Klink, Carlo D'Eramo, Jan Peters et al.

Curriculum reinforcement learning (CRL) allows solving complex tasks by generating a tailored sequence of learning tasks, starting from easy ones and subsequently increasing their difficulty. Although the potential of curricula in RL has been clearly shown in various works, it is less clear how to generate them for a given learning environment, resulting in various methods aiming to automate this task. In this work, we focus on framing curricula as interpolations between task distributions, which has previously been shown to be a viable approach to CRL. Identifying key issues of existing methods, we frame the generation of a curriculum as a constrained optimal transport problem between task distributions. Benchmarks show that this way of curriculum generation can improve upon existing CRL methods, yielding high performance in various tasks with different characteristics.

Changling Li, Zhang-Wei Hong, Pulkit Agrawal et al.

Experience replay serves as a key component in the success of online reinforcement learning (RL). Prioritized experience replay (PER) reweights experiences by the temporal difference (TD) error empirically enhancing the performance. However, few works have explored the motivation of using TD error. In this work, we provide an alternative perspective on TD-error-based reweighting. We show the connections between the experience prioritization and occupancy optimization. By using a regularized RL objective with $f-$divergence regularizer and employing its dual form, we show that an optimal solution to the objective is obtained by shifting the distribution of off-policy data in the replay buffer towards the on-policy optimal distribution using TD-error-based occupancy ratios. Our derivation results in a new pipeline of TD error prioritization. We specifically explore the KL divergence as the regularizer and obtain a new form of prioritization scheme, the regularized optimal experience replay (ROER). We evaluate the proposed prioritization scheme with the Soft Actor-Critic (SAC) algorithm in continuous control MuJoCo and DM Control benchmark tasks where our proposed scheme outperforms baselines in 6 out of 11 tasks while the results of the rest match with or do not deviate far from the baselines. Further, using pretraining, ROER achieves noticeable improvement on difficult Antmaze environment where baselines fail, showing applicability to offline-to-online fine-tuning. Code is available at \url{https://github.com/XavierChanglingLi/Regularized-Optimal-Experience-Replay}.

Yi Zhao, Wenshuai Zhao, Rinu Boney et al.

Reinforcement learning is able to solve complex sequential decision-making tasks but is currently limited by sample efficiency and required computation. To improve sample efficiency, recent work focuses on model-based RL which interleaves model learning with planning. Recent methods further utilize policy learning, value estimation, and, self-supervised learning as auxiliary objectives. In this paper we show that, surprisingly, a simple representation learning approach relying only on a latent dynamics model trained by latent temporal consistency is sufficient for high-performance RL. This applies when using pure planning with a dynamics model conditioned on the representation, but, also when utilizing the representation as policy and value function features in model-free RL. In experiments, our approach learns an accurate dynamics model to solve challenging high-dimensional locomotion tasks with online planners while being 4.1 times faster to train compared to ensemble-based methods. With model-free RL without planning, especially on high-dimensional tasks, such as the DeepMind Control Suite Humanoid and Dog tasks, our approach outperforms model-free methods by a large margin and matches model-based methods' sample efficiency while training 2.4 times faster.

Pascal Klink, Florian Wolf, Kai Ploeger et al.

Reinforcement Learning (RL) allows learning non-trivial robot control laws purely from data. However, many successful applications of RL have relied on ad-hoc regularizations, such as hand-crafted curricula, to regularize the learning performance. In this paper, we pair a recent algorithm for automatically building curricula with RL on massively parallelized simulations to learn a tracking controller for a spherical pendulum on a robotic arm via RL. Through an improved optimization scheme that better respects the non-Euclidean task structure, we allow the method to reliably generate curricula of trajectories to be tracked, resulting in faster and more robust learning compared to an RL baseline that does not exploit this form of structured learning. The learned policy matches the performance of an optimal control baseline on the real system, demonstrating the potential of curriculum RL to jointly learn state estimation and control for non-linear tracking tasks.

Yi Zhao, Le Chen, Jan Schneider et al.

It has been a long-standing research goal to endow robot hands with human-level dexterity. Bi-manual robot piano playing constitutes a task that combines challenges from dynamic tasks, such as generating fast while precise motions, with slower but contact-rich manipulation problems. Although reinforcement learning based approaches have shown promising results in single-task performance, these methods struggle in a multi-song setting. Our work aims to close this gap and, thereby, enable imitation learning approaches for robot piano playing at scale. To this end, we introduce the Robot Piano 1 Million (RP1M) dataset, containing bi-manual robot piano playing motion data of more than one million trajectories. We formulate finger placements as an optimal transport problem, thus, enabling automatic annotation of vast amounts of unlabeled songs. Benchmarking existing imitation learning approaches shows that such approaches reach state-of-the-art robot piano playing performance by leveraging RP1M.

Wenshuai Zhao, Yi Zhao, Zhiyuan Li et al.

*Relative overgeneralization* (RO) occurs in cooperative multi-agent learning tasks when agents converge towards a suboptimal joint policy due to overfitting to suboptimal behavior of other agents. No methods have been proposed for addressing RO in multi-agent policy gradient (MAPG) methods although these methods produce state-of-the-art results. To address this gap, we propose a general, yet simple, framework to enable optimistic updates in MAPG methods that alleviate the RO problem. Our approach involves clipping the advantage to eliminate negative values, thereby facilitating optimistic updates in MAPG. The optimism prevents individual agents from quickly converging to a local optimum. Additionally, we provide a formal analysis to show that the proposed method retains optimality at a fixed point. In extensive evaluations on a diverse set of tasks including the *Multi-agent MuJoCo* and *Overcooked* benchmarks, our method outperforms strong baselines on 13 out of 19 tested tasks and matches the performance on the rest.

Kalle Kujanpää, Amin Babadi, Yi Zhao et al.

Online planning is crucial for high performance in many complex sequential decision-making tasks. Monte Carlo Tree Search (MCTS) employs a principled mechanism for trading off exploration for exploitation for efficient online planning, and it outperforms comparison methods in many discrete decision-making domains such as Go, Chess, and Shogi. Subsequently, extensions of MCTS to continuous domains have been developed. However, the inherent high branching factor and the resulting explosion of the search tree size are limiting the existing methods. To address this problem, we propose Continuous Monte Carlo Graph Search (CMCGS), an extension of MCTS to online planning in environments with continuous state and action spaces. CMCGS takes advantage of the insight that, during planning, sharing the same action policy between several states can yield high performance. To implement this idea, at each time step, CMCGS clusters similar states into a limited number of stochastic action bandit nodes, which produce a layered directed graph instead of an MCTS search tree. Experimental evaluation shows that CMCGS outperforms comparable planning methods in several complex continuous DeepMind Control Suite benchmarks and 2D navigation and exploration tasks with limited sample budgets. Furthermore, CMCGS can be scaled up through parallelization, and it outperforms the Cross-Entropy Method (CEM) in continuous control with learned dynamics models.

Rongzhen Zhao, Vivienne Wang, Juho Kannala et al.

Similar to humans perceiving visual scenes as objects, Object-Centric Learning (OCL) can abstract dense images or videos into sparse object-level features. Transformer-based OCL handles complex textures well due to the decoding guidance of discrete representation, obtained by discretizing noisy features in image or video feature maps using template features from a codebook. However, treating features as minimal units overlooks their composing attributes, thus impeding model generalization; indexing features with natural numbers loses attribute-level commonalities and characteristics, thus diminishing heuristics for model convergence. We propose \textit{Grouped Discrete Representation} (GDR) to address these issues by grouping features into attributes and indexing them with tuple numbers. In extensive experiments across different query initializations, dataset modalities, and model architectures, GDR consistently improves convergence and generalizability. Visualizations show that our method effectively captures attribute-level information in features. The source code will be available upon acceptance.

Kalle Kujanpää, Joni Pajarinen, Alexander Ilin

The ability to plan actions on multiple levels of abstraction enables intelligent agents to solve complex tasks effectively. However, learning the models for both low and high-level planning from demonstrations has proven challenging, especially with higher-dimensional inputs. To address this issue, we propose to use reinforcement learning to identify subgoals in expert trajectories by associating the magnitude of the rewards with the predictability of low-level actions given the state and the chosen subgoal. We build a vector-quantized generative model for the identified subgoals to perform subgoal-level planning. In experiments, the algorithm excels at solving complex, long-horizon decision-making problems outperforming state-of-the-art. Because of its ability to plan, our algorithm can find better trajectories than the ones in the training set

Kalle Kujanpää, Joni Pajarinen, Alexander Ilin

Solving complex planning problems has been a long-standing challenge in computer science. Learning-based subgoal search methods have shown promise in tackling these problems, but they often suffer from a lack of completeness guarantees, meaning that they may fail to find a solution even if one exists. In this paper, we propose an efficient approach to augment a subgoal search method to achieve completeness in discrete action spaces. Specifically, we augment the high-level search with low-level actions to execute a multi-level (hybrid) search, which we call complete subgoal search. This solution achieves the best of both worlds: the practical efficiency of high-level search and the completeness of low-level search. We apply the proposed search method to a recently proposed subgoal search algorithm and evaluate the algorithm trained on offline data on complex planning problems. We demonstrate that our complete subgoal search not only guarantees completeness but can even improve performance in terms of search expansions for instances that the high-level could solve without low-level augmentations. Our approach makes it possible to apply subgoal-level planning for systems where completeness is a critical requirement.

Tuan Dam, Pascal Stenger, Lukas Schneider et al.

This paper introduces a novel backup strategy for Monte-Carlo Tree Search (MCTS) designed for highly stochastic and partially observable Markov decision processes. We adopt a probabilistic approach, modeling both value and action-value nodes as Gaussian distributions. We introduce a novel backup operator that computes value nodes as the Wasserstein barycenter of their action-value children nodes; thus, propagating the uncertainty of the estimate across the tree to the root node. We study our novel backup operator when using a novel combination of $L^1$-Wasserstein barycenter with $α$-divergence, by drawing a notable connection to the generalized mean backup operator. We complement our probabilistic backup operator with two sampling strategies, based on optimistic selection and Thompson sampling, obtaining our Wasserstein MCTS algorithm. We provide theoretical guarantees of asymptotic convergence to the optimal policy, and an empirical evaluation on several stochastic and partially observable environments, where our approach outperforms well-known related baselines.

Aidan Scannell, Riccardo Mereu, Paul Chang et al.

Deep neural networks (NNs) are known to lack uncertainty estimates and struggle to incorporate new data. We present a method that mitigates these issues by converting NNs from weight space to function space, via a dual parameterization. Importantly, the dual parameterization enables us to formulate a sparse representation that captures information from the entire data set. This offers a compact and principled way of capturing uncertainty and enables us to incorporate new data without retraining whilst retaining predictive performance. We provide proof-of-concept demonstrations with the proposed approach for quantifying uncertainty in supervised learning on UCI benchmark tasks.

Wenyan Yang, Huiling Wang, Dingding Cai et al.

Offline goal-conditioned reinforcement learning (GCRL) can be challenging due to overfitting to the given dataset. To generalize agents' skills outside the given dataset, we propose a goal-swapping procedure that generates additional trajectories. To alleviate the problem of noise and extrapolation errors, we present a general offline reinforcement learning method called deterministic Q-advantage policy gradient (DQAPG). In the experiments, DQAPG outperforms state-of-the-art goal-conditioned offline RL methods in a wide range of benchmark tasks, and goal-swapping further improves the test results. It is noteworthy, that the proposed method obtains good performance on the challenging dexterous in-hand manipulation tasks for which the prior methods failed.

Yihao Wang, Yang Miao, Wenshuai Zhao et al.

Articulation perception aims to recover the motion and structure of articulated objects (e.g., drawers and cupboards), and is fundamental to 3D scene understanding in robotics, simulation, and animation. Existing learning-based methods rely heavily on supervised training with high-quality 3D data and manual annotations, limiting scalability and diversity. To address this limitation, we propose PAWS, a method that directly extracts object articulations from hand-object interactions in large-scale in-the-wild egocentric videos. We evaluate our method on the public data sets, including HD-EPIC and Arti4D data sets, achieving significant improvements over baselines. We further demonstrate that the extracted articulations benefit downstream tasks, including fine-tuning 3D articulation prediction models and enabling robot manipulation. See the project website at https://aaltoml.github.io/PAWS/.

Wenyan Yang, Joni Pajarinen, Dinging Cai et al.

In goal-conditioned offline reinforcement learning, an agent learns from previously collected data to go to an arbitrary goal. Since the offline data only contains a finite number of trajectories, a main challenge is how to generate more data. Goal-swapping generates additional data by switching trajectory goals but while doing so produces a large number of invalid trajectories. To address this issue, we propose prioritized goal-swapping experience replay (PGSER). PGSER uses a pre-trained Q function to assign higher priority weights to goal swapped transitions that allow reaching the goal. In experiments, PGSER significantly improves over baselines in a wide range of benchmark tasks, including challenging previously unsuccessful dexterous in-hand manipulation tasks.

Aidan Scannell, Riccardo Mereu, Paul Chang et al.

Sequential learning paradigms pose challenges for gradient-based deep learning due to difficulties incorporating new data and retaining prior knowledge. While Gaussian processes elegantly tackle these problems, they struggle with scalability and handling rich inputs, such as images. To address these issues, we introduce a technique that converts neural networks from weight space to function space, through a dual parameterization. Our parameterization offers: (i) a way to scale function-space methods to large data sets via sparsification, (ii) retention of prior knowledge when access to past data is limited, and (iii) a mechanism to incorporate new data without retraining. Our experiments demonstrate that we can retain knowledge in continual learning and incorporate new data efficiently. We further show its strengths in uncertainty quantification and guiding exploration in model-based RL. Further information and code is available on the project website.

Rongzhen Zhao, Vivienne Wang, Juho Kannala et al.

Object-Centric Learning (OCL) aims to discover objects in images or videos by reconstructing the input. Representative methods achieve this by reconstructing the input as its Variational Autoencoder (VAE) discrete representations, which suppress (super-)pixel noise and enhance object separability. However, these methods treat features as indivisible units, overlooking their compositional attributes, and discretize features via scalar code indexes, losing attribute-level similarities and differences. We propose Grouped Discrete Representation (GDR) for OCL. For better generalization, features are decomposed into combinatorial attributes by organized channel grouping. For better convergence, features are quantized into discrete representations via tuple code indexes. Experiments demonstrate that GDR consistently improves both mainstream and state-of-the-art OCL methods across various datasets. Visualizations further highlight GDR's superior object separability and interpretability. The source code is available on https://github.com/Genera1Z/GroupedDiscreteRepresentation.

Rongzhen Zhao, Yi Zhao, Juho Kannala et al.

Unlike popular solutions based on dense feature maps, Object-Centric Learning (OCL) represents visual scenes as sub-symbolic object-level feature vectors, termed slots, which are highly versatile for tasks involving visual modalities. OCL typically aggregates object superpixels into slots by iteratively applying competitive cross attention, known as Slot Attention, with the slots as the query. However, once initialized, these slots are reused naively, causing redundant slots to compete with informative ones for representing objects. This often results in objects being erroneously segmented into parts. Additionally, mainstream methods derive supervision signals solely from decoding slots into the input's reconstruction, overlooking potential supervision based on internal information. To address these issues, we propose Slot Attention with re-Initialization and self-Distillation (DIAS): $\emph{i)}$ We reduce redundancy in the aggregated slots and re-initialize extra aggregation to update the remaining slots; $\emph{ii)}$ We drive the bad attention map at the first aggregation iteration to approximate the good at the last iteration to enable self-distillation. Experiments demonstrate that DIAS achieves state-of-the-art on OCL tasks like object discovery and recognition, while also improving advanced visual prediction and reasoning. Our source code and model checkpoints are available on https://github.com/Genera1Z/DIAS.

Rongzhen Zhao, Vivienne Wang, Juho Kannala et al.

Object-Centric Learning (OCL) aggregates image or video feature maps into object-level feature vectors, termed \textit{slots}. It's self-supervision of reconstructing the input from slots struggles with complex object textures, thus Vision Foundation Model (VFM) representations are used as the aggregation input and reconstruction target. Existing methods leverage VFM representations in diverse ways yet fail to fully exploit their potential. In response, we propose a unified architecture, Vector-Quantized VFMs for OCL (VQ-VFM-OCL, or VVO). The key to our unification is simply shared quantizing VFM representations in OCL aggregation and decoding. Experiments show that across different VFMs, aggregators and decoders, our VVO consistently outperforms baselines in object discovery and recognition, as well as downstream visual prediction and reasoning. We also mathematically analyze why VFM representations facilitate OCL aggregation and why their shared quantization as reconstruction targets strengthens OCL supervision. Our source code and model checkpoints are available on https://github.com/Genera1Z/VQ-VFM-OCL.

Rongzhen Zhao, Jian Li, Juho Kannala et al.

Unsupervised video Object-Centric Learning (OCL) is promising as it enables object-level scene representation and dynamics modeling as we humans do. Mainstream video OCL methods adopt a recurrent architecture: An aggregator aggregates current video frame into object features, termed slots, under some queries; A transitioner transits current slots to queries for the next frame. This is an effective architecture but all existing implementations both (\textit{i1}) neglect to incorporate next frame features, the most informative source for query prediction, and (\textit{i2}) fail to learn transition dynamics, the knowledge essential for query prediction. To address these issues, we propose Random Slot-Feature pair for learning Query prediction (RandSF.Q): (\textit{t1}) We design a new transitioner to incorporate both slots and features, which provides more information for query prediction; (\textit{t2}) We train the transitioner to predict queries from slot-feature pairs randomly sampled from available recurrences, which drives it to learn transition dynamics. Experiments on scene representation demonstrate that our method surpass existing video OCL methods significantly, e.g., up to 10 points on object discovery, setting new state-of-the-art. Such superiority also benefits downstream tasks like dynamics modeling. Our core source code, model checkpoints and training logs are available on https://github.com/Genera1Z/RandSF.Q.

Simone Parisi, Davide Tateo, Maximilian Hensel et al.

Reinforcement learning with sparse rewards is still an open challenge. Classic methods rely on getting feedback via extrinsic rewards to train the agent, and in situations where this occurs very rarely the agent learns slowly or cannot learn at all. Similarly, if the agent receives also rewards that create suboptimal modes of the objective function, it will likely prematurely stop exploring. More recent methods add auxiliary intrinsic rewards to encourage exploration. However, auxiliary rewards lead to a non-stationary target for the Q-function. In this paper, we present a novel approach that (1) plans exploration actions far into the future by using a long-term visitation count, and (2) decouples exploration and exploitation by learning a separate function assessing the exploration value of the actions. Contrary to existing methods which use models of reward and dynamics, our approach is off-policy and model-free. We further propose new tabular environments for benchmarking exploration in reinforcement learning. Empirical results on classic and novel benchmarks show that the proposed approach outperforms existing methods in environments with sparse rewards, especially in the presence of rewards that create suboptimal modes of the objective function. Results also suggest that our approach scales gracefully with the size of the environment. Source code is available at https://github.com/sparisi/visit-value-explore

Hongjia Liu, Rongzhen Zhao, Haohan Chen et al.

Learning object-level, structured representations is widely regarded as a key to better generalization in vision and underpins the design of next-generation Pre-trained Vision Models (PVMs). Mainstream Object-Centric Learning (OCL) methods adopt Slot Attention or its variants to iteratively aggregate objects' super-pixels into a fixed set of query feature vectors, termed slots. However, their reliance on a static slot count leads to an object being represented as multiple parts when the number of objects varies. We introduce MetaSlot, a plug-and-play Slot Attention variant that adapts to variable object counts. MetaSlot (i) maintains a codebook that holds prototypes of objects in a dataset by vector-quantizing the resulting slot representations; (ii) removes duplicate slots from the traditionally aggregated slots by quantizing them with the codebook; and (iii) injects progressively weaker noise into the Slot Attention iterations to accelerate and stabilize the aggregation. MetaSlot is a general Slot Attention variant that can be seamlessly integrated into existing OCL architectures. Across multiple public datasets and tasks--including object discovery and recognition--models equipped with MetaSlot achieve significant performance gains and markedly interpretable slot representations, compared with existing Slot Attention variants.

Yi Zhao, Aidan Scannell, Wenshuai Zhao et al.

Leveraging offline data is a promising way to improve the sample efficiency of online reinforcement learning (RL). This paper expands the pool of usable data for offline-to-online RL by leveraging abundant non-curated data that is reward-free, of mixed quality, and collected across multiple embodiments. Although learning a world model appears promising for utilizing such data, we find that naive fine-tuning fails to accelerate RL training on many tasks. Through careful investigation, we attribute this failure to the distributional shift between offline and online data during fine-tuning. To address this issue and effectively use the offline data, we propose two essential techniques: \emph{i)} experience rehearsal and \emph{ii)} execution guidance. With these modifications, the non-curated offline data substantially improves RL's sample efficiency. Under limited sample budgets, our method achieves a 102.8\% relative improvement in aggregate score over learning-from-scratch baselines across 72 visuomotor tasks spanning 6 embodiments. On challenging tasks such as locomotion and robotic manipulation, it outperforms prior methods that utilize offline data by a decent margin.

Yongxi Cao, Julian F. Schumann, Jens Kober et al.

Deep generative models such as conditional variational autoencoders (CVAEs) have shown great promise for predicting trajectories of surrounding agents in autonomous vehicle planning. State-of-the-art models have achieved remarkable accuracy in such prediction tasks. Besides accuracy, diversity is also crucial for safe planning because human behaviors are inherently uncertain and multimodal. However, existing methods generally lack a scheme to generate controllably diverse trajectories, which is arguably more useful than randomly diversified trajectories, to the end of safe planning. To address this, we propose PrefCVAE, an augmented CVAE framework that uses weakly labeled preference pairs to imbue latent variables with semantic attributes. Using average velocity as an example attribute, we demonstrate that PrefCVAE enables controllable, semantically meaningful predictions without degrading baseline accuracy. Our results show the effectiveness of preference supervision as a cost-effective way to enhance sampling-based generative models.

Xing Lei, Wenyan Yang, Kaiqiang Ke et al.

Goal-conditioned reinforcement learning (GCRL) with sparse rewards remains a fundamental challenge in reinforcement learning. While hindsight experience replay (HER) has shown promise by relabeling collected trajectories with achieved goals, we argue that trajectory relabeling alone does not fully exploit the available experiences in off-policy GCRL methods, resulting in limited sample efficiency. In this paper, we propose Hindsight Goal-conditioned Regularization (HGR), a technique that generates action regularization priors based on hindsight goals. When combined with hindsight self-imitation regularization (HSR), our approach enables off-policy RL algorithms to maximize experience utilization. Compared to existing GCRL methods that employ HER and self-imitation techniques, our hindsight regularizations achieve substantially more efficient sample reuse and the best performances, which we empirically demonstrate on a suite of navigation and manipulation tasks.

Vivienne Huiling Wang, Tinghuai Wang, Joni Pajarinen

Hierarchical reinforcement learning (HRL) learns to make decisions on multiple levels of temporal abstraction. A key challenge in HRL is that the low-level policy changes over time, making it difficult for the high-level policy to generate effective subgoals. To address this issue, the high-level policy must capture a complex subgoal distribution while also accounting for uncertainty in its estimates. We propose an approach that trains a conditional diffusion model regularized by a Gaussian Process (GP) prior to generate a complex variety of subgoals while leveraging principled GP uncertainty quantification. Building on this framework, we develop a strategy that selects subgoals from both the diffusion policy and GP's predictive mean. Our approach outperforms prior HRL methods in both sample efficiency and performance on challenging continuous control benchmarks.

Wenyan Yang, Joni Pajarinen

Offline goal-conditioned reinforcement learning (RL) relies on fixed datasets where many potential goals share the same state and action spaces. However, these potential goals are not explicitly represented in the collected trajectories. To learn a generalizable goal-conditioned policy, it is common to sample goals and state-action pairs uniformly using dynamic programming methods such as Q-learning. Uniform sampling, however, requires an intractably large dataset to cover all possible combinations and creates many unreachable state-goal-action pairs that degrade policy performance. Our key insight is that sampling should favor transitions that enable goal achievement. To this end, we propose Reachability Weighted Sampling (RWS). RWS uses a reachability classifier trained via positive-unlabeled (PU) learning on goal-conditioned state-action values. The classifier maps these values to a reachability score, which is then used as a sampling priority. RWS is a plug-and-play module that integrates seamlessly with standard offline RL algorithms. Experiments on six complex simulated robotic manipulation tasks, including those with a robot arm and a dexterous hand, show that RWS significantly improves performance. In one notable case, performance on the HandBlock-Z task improved by nearly 50 percent relative to the baseline. These results indicate the effectiveness of reachability-weighted sampling.

Mehdi Moshtaghi, Siavash H. Khajavi, Joni Pajarinen

We introduce RGB-Th-Bench, the first benchmark designed to evaluate the ability of Vision-Language Models (VLMs) to comprehend RGB-Thermal image pairs. While VLMs have demonstrated remarkable progress in visual reasoning and multimodal understanding, their evaluation has been predominantly limited to RGB-based benchmarks, leaving a critical gap in assessing their capabilities in infrared vision tasks. Existing visible-infrared datasets are either task-specific or lack high-quality annotations necessary for rigorous model evaluation. To address these limitations, RGB-Th-Bench provides a comprehensive evaluation framework covering 14 distinct skill dimensions, with a total of 1,600+ expert-annotated Yes/No questions. The benchmark employs two accuracy metrics: a standard question-level accuracy and a stricter skill-level accuracy, which evaluates model robustness across multiple questions within each skill dimension. This design ensures a thorough assessment of model performance, including resilience to adversarial and hallucinated responses. We conduct extensive evaluations on 19 state-of-the-art VLMs, revealing significant performance gaps in RGB-Thermal understanding. Our results show that even the strongest models struggle with thermal image comprehension, with performance heavily constrained by their RGB-based capabilities. Additionally, the lack of large-scale application-specific and expert-annotated thermal-caption-pair datasets in pre-training is an important reason of the observed performance gap. RGB-Th-Bench highlights the urgent need for further advancements in multimodal learning to bridge the gap between visible and thermal image understanding. The dataset is available through this link, and the evaluation code will also be made publicly available.

Mohammadreza Nakhaei, Aidan Scannell, Joni Pajarinen

Offline meta-reinforcement learning aims to equip agents with the ability to rapidly adapt to new tasks by training on data from a set of different tasks. Context-based approaches utilize a history of state-action-reward transitions -- referred to as the context -- to infer representations of the current task, and then condition the agent, i.e., the policy and value function, on the task representations. Intuitively, the better the task representations capture the underlying tasks, the better the agent can generalize to new tasks. Unfortunately, context-based approaches suffer from distribution mismatch, as the context in the offline data does not match the context at test time, limiting their ability to generalize to the test tasks. This leads to the task representations overfitting to the offline training data. Intuitively, the task representations should be independent of the behavior policy used to collect the offline data. To address this issue, we approximately minimize the mutual information between the distribution over the task representations and behavior policy by maximizing the entropy of behavior policy conditioned on the task representations. We validate our approach in MuJoCo environments, showing that compared to baselines, our task representations more faithfully represent the underlying tasks, leading to outperforming prior methods in both in-distribution and out-of-distribution tasks.

Vivienne Huiling Wang, Tinghuai Wang, Wenyan Yang et al.

In goal-conditioned hierarchical reinforcement learning (HRL), a high-level policy specifies a subgoal for the low-level policy to reach. Effective HRL hinges on a suitable subgoal represen tation function, abstracting state space into latent subgoal space and inducing varied low-level behaviors. Existing methods adopt a subgoal representation that provides a deterministic mapping from state space to latent subgoal space. Instead, this paper utilizes Gaussian Processes (GPs) for the first probabilistic subgoal representation. Our method employs a GP prior on the latent subgoal space to learn a posterior distribution over the subgoal representation functions while exploiting the long-range correlation in the state space through learnable kernels. This enables an adaptive memory that integrates long-range subgoal information from prior planning steps allowing to cope with stochastic uncertainties. Furthermore, we propose a novel learning objective to facilitate the simultaneous learning of probabilistic subgoal representations and policies within a unified framework. In experiments, our approach outperforms state-of-the-art baselines in standard benchmarks but also in environments with stochastic elements and under diverse reward conditions. Additionally, our model shows promising capabilities in transferring low-level policies across different tasks.

Mohammadreza Nakhaei, Aidan Scannell, Joni Pajarinen

Offline reinforcement learning (RL) allows learning sequential behavior from fixed datasets. Since offline datasets do not cover all possible situations, many methods collect additional data during online fine-tuning to improve performance. In general, these methods assume that the transition dynamics remain the same during both the offline and online phases of training. However, in many real-world applications, such as outdoor construction and navigation over rough terrain, it is common for the transition dynamics to vary between the offline and online phases. Moreover, the dynamics may vary during the online fine-tuning. To address this problem of changing dynamics from offline to online RL we propose a residual learning approach that infers dynamics changes to correct the outputs of the offline solution. At the online fine-tuning phase, we train a context encoder to learn a representation that is consistent inside the current online learning environment while being able to predict dynamic transitions. Experiments in D4RL MuJoCo environments, modified to support dynamics' changes upon environment resets, show that our approach can adapt to these dynamic changes and generalize to unseen perturbations in a sample-efficient way, whilst comparison methods cannot.

Aidan Scannell, Kalle Kujanpää, Yi Zhao et al.

Learning representations for reinforcement learning (RL) has shown much promise for continuous control. We propose an efficient representation learning method using only a self-supervised latent-state consistency loss. Our approach employs an encoder and a dynamics model to map observations to latent states and predict future latent states, respectively. We achieve high performance and prevent representation collapse by quantizing the latent representation such that the rank of the representation is empirically preserved. Our method, named iQRL: implicitly Quantized Reinforcement Learning, is straightforward, compatible with any model-free RL algorithm, and demonstrates excellent performance by outperforming other recently proposed representation learning methods in continuous control benchmarks from DeepMind Control Suite.

Zhiyuan Li, Wenshuai Zhao, Lijun Wu et al.

In multi-agent reinforcement learning, centralized training with decentralized execution (CTDE) methods typically assume that agents make decisions based on their local observations independently, which may not lead to a correlated joint policy with coordination. Coordination can be explicitly encouraged during training and individual policies can be trained to imitate the correlated joint policy. However, this may lead to an \textit{asymmetric learning failure} due to the observation mismatch between the joint and individual policies. Inspired by the concept of correlated equilibrium, we introduce a \textit{strategy modification} called AgentMixer that allows agents to correlate their policies. AgentMixer combines individual partially observable policies into a joint fully observable policy non-linearly. To enable decentralized execution, we introduce \textit{Individual-Global-Consistency} to guarantee mode consistency during joint training of the centralized and decentralized policies and prove that AgentMixer converges to an $ε$-approximate Correlated Equilibrium. In the Multi-Agent MuJoCo, SMAC-v2, Matrix Game, and Predator-Prey benchmarks, AgentMixer outperforms or matches state-of-the-art methods.

Zhang-Wei Hong, Tao Chen, Yen-Chen Lin et al.

State-of-the-art deep Q-learning methods update Q-values using state transition tuples sampled from the experience replay buffer. This strategy often uniformly and randomly samples or prioritizes data sampling based on measures such as the temporal difference (TD) error. Such sampling strategies can be inefficient at learning Q-function because a state's Q-value depends on the Q-value of successor states. If the data sampling strategy ignores the precision of the Q-value estimate of the next state, it can lead to useless and often incorrect updates to the Q-values. To mitigate this issue, we organize the agent's experience into a graph that explicitly tracks the dependency between Q-values of states. Each edge in the graph represents a transition between two states by executing a single action. We perform value backups via a breadth-first search starting from that expands vertices in the graph starting from the set of terminal states and successively moving backward. We empirically show that our method is substantially more data-efficient than several baselines on a diverse range of goal-reaching tasks. Notably, the proposed method also outperforms baselines that consume more batches of training experience and operates from high-dimensional observational data such as images.

Abdolreza Taheri, Joni Pajarinen, Reza Ghabcheloo

The ability of Gaussian processes (GPs) to predict the behavior of dynamical systems as a more sample-efficient alternative to parametric models seems promising for real-world robotics research. However, the computational complexity of GPs has made policy search a highly time and memory consuming process that has not been able to scale to larger problems. In this work, we develop a policy optimization method by leveraging fast predictive sampling methods to process batches of trajectories in every forward pass, and compute gradient updates over policy parameters by automatic differentiation of Monte Carlo evaluations, all on GPU. We demonstrate the effectiveness of our approach in training policies on a set of reference-tracking control experiments with a heavy-duty machine. Benchmark results show a significant speedup over exact methods and showcase the scalability of our method to larger policy networks, longer horizons, and up to thousands of trajectories with a sublinear drop in speed.

Tuan Dam, Carlo D'Eramo, Jan Peters et al.

Monte-Carlo Tree Search (MCTS) is a class of methods for solving complex decision-making problems through the synergy of Monte-Carlo planning and Reinforcement Learning (RL). The highly combinatorial nature of the problems commonly addressed by MCTS requires the use of efficient exploration strategies for navigating the planning tree and quickly convergent value backup methods. These crucial problems are particularly evident in recent advances that combine MCTS with deep neural networks for function approximation. In this work, we propose two methods for improving the convergence rate and exploration based on a newly introduced backup operator and entropy regularization. We provide strong theoretical guarantees to bound convergence rate, approximation error, and regret of our methods. Moreover, we introduce a mathematical framework based on the use of the $α$-divergence for backup and exploration in MCTS. We show that this theoretical formulation unifies different approaches, including our newly introduced ones, under the same mathematical framework, allowing to obtain different methods by simply changing the value of $α$. In practice, our unified perspective offers a flexible way to balance between exploration and exploitation by tuning the single $α$ parameter according to the problem at hand. We validate our methods through a rigorous empirical study from basic toy problems to the complex Atari games, and including both MDP and POMDP problems.

Vivienne Huiling Wang, Joni Pajarinen, Tinghuai Wang et al.

Hierarchical reinforcement learning (HRL) proposes to solve difficult tasks by performing decision-making and control at successively higher levels of temporal abstraction. However, off-policy HRL often suffers from the problem of a non-stationary high-level policy since the low-level policy is constantly changing. In this paper, we propose a novel HRL approach for mitigating the non-stationarity by adversarially enforcing the high-level policy to generate subgoals compatible with the current instantiation of the low-level policy. In practice, the adversarial learning is implemented by training a simple state-conditioned discriminator network concurrently with the high-level policy which determines the compatibility level of subgoals. Comparison to state-of-the-art algorithms shows that our approach improves both learning efficiency and performance in challenging continuous control tasks.

Stephan Weigand, Pascal Klink, Jan Peters et al.

Due to recent breakthroughs, reinforcement learning (RL) has demonstrated impressive performance in challenging sequential decision-making problems. However, an open question is how to make RL cope with partial observability which is prevalent in many real-world problems. Contrary to contemporary RL approaches, which focus mostly on improved memory representations or strong assumptions about the type of partial observability, we propose a simple but efficient approach that can be applied together with a wide variety of RL methods. Our main insight is that smoothly transitioning from full observability to partial observability during the training process yields a high performance policy. The approach, called partially observable guided reinforcement learning (PO-GRL), allows to utilize full state information during policy optimization without compromising the optimality of the final policy. A comprehensive evaluation in discrete partially observableMarkov decision process (POMDP) benchmark problems and continuous partially observable MuJoCo and OpenAI gym tasks shows that PO-GRL improves performance. Finally, we demonstrate PO-GRL in the ball-in-the-cup task on a real Barrett WAM robot under partial observability.

Wenyan Yang, Nataliya Strokina, Nikolay Serbenyuk et al.

We have recently proposed two pile loading controllers that learn from human demonstrations: a neural network (NNet) [1] and a random forest (RF) controller [2]. In the field experiments the RF controller obtained clearly better success rates. In this work, the previous findings are drastically revised by experimenting summer time trained controllers in winter conditions. The winter experiments revealed a need for additional sensors, more training data, and a controller that can take advantage of these. Therefore, we propose a revised neural controller (NNetV2) which has a more expressive structure and uses a neural attention mechanism to focus on important parts of the sensor and control signals. Using the same data and sensors to train and test the three controllers, NNetV2 achieves better robustness against drastically changing conditions and superior success rate. To the best of our knowledge, this is the first work testing a learning-based controller for a heavy-duty machine in drastically varying outdoor conditions and delivering high success rate in winter, being trained in summer.

Pascal Klink, Hany Abdulsamad, Boris Belousov et al.

Across machine learning, the use of curricula has shown strong empirical potential to improve learning from data by avoiding local optima of training objectives. For reinforcement learning (RL), curricula are especially interesting, as the underlying optimization has a strong tendency to get stuck in local optima due to the exploration-exploitation trade-off. Recently, a number of approaches for an automatic generation of curricula for RL have been shown to increase performance while requiring less expert knowledge compared to manually designed curricula. However, these approaches are seldomly investigated from a theoretical perspective, preventing a deeper understanding of their mechanics. In this paper, we present an approach for automated curriculum generation in RL with a clear theoretical underpinning. More precisely, we formalize the well-known self-paced learning paradigm as inducing a distribution over training tasks, which trades off between task complexity and the objective to match a desired task distribution. Experiments show that training on this induced distribution helps to avoid poor local optima across RL algorithms in different tasks with uninformative rewards and challenging exploration requirements.

Joni Pajarinen, Jens Lundell, Ville Kyrki

Manipulating unknown objects in a cluttered environment is difficult because segmentation of the scene into objects, that is, object composition is uncertain. Due to this uncertainty, earlier work has concentrated on either identifying the "best" object composition and deciding on manipulation actions accordingly, or, tried to greedily gather information about the "best" object composition. Contrary to earlier work, we 1) utilize different possible object compositions in planning, 2) take advantage of object composition information provided by robot actions, 3) take into account the effect of different competing object hypotheses on the actual task to be performed. We cast the manipulation planning problem as a partially observable Markov decision process (POMDP) which plans over possible hypotheses of object compositions. The POMDP model chooses the action that maximizes the long-term expected task specific utility, and while doing so, considers the value of informative actions and the effect of different object hypotheses on the completion of the task. In simulations and in experiments with an RGB-D sensor, a Kinova Jaco and a Franka Emika Panda robot arm, a probabilistic approach outperforms an approach that only considers the most likely object composition and long term planning outperforms greedy decision making.

Joni Pajarinen

Optimizing a partially observable Markov decision process (POMDP) policy is challenging. The policy graph improvement (PGI) algorithm for POMDPs represents the policy as a fixed size policy graph and improves the policy monotonically. Due to the fixed policy size, computation time for each improvement iteration is known in advance. Moreover, the method allows for compact understandable policies. This report describes the technical details of the PGI [1] and particle based PGI [2] algorithms for POMDPs in a more accessible way than [1] or [2] allowing practitioners and students to understand and implement the algorithms.

Mikko Lauri, Joni Pajarinen, Jan Peters et al.

3D scene models are useful in robotics for tasks such as path planning, object manipulation, and structural inspection. We consider the problem of creating a 3D model using depth images captured by a team of multiple robots. Each robot selects a viewpoint and captures a depth image from it, and the images are fused to update the scene model. The process is repeated until a scene model of desired quality is obtained. Next-best-view planning uses the current scene model to select the next viewpoints. The objective is to select viewpoints so that the images captured using them improve the quality of the scene model the most. In this paper, we address next-best-view planning for multiple depth cameras. We propose a utility function that scores sets of viewpoints and avoids overlap between multiple sensors. We show that multi-sensor next-best-view planning with this utility function is an instance of submodular maximization under a matroid constraint. This allows the planning problem to be solved by a polynomial-time greedy algorithm that yields a solution within a constant factor from the optimal. We evaluate the performance of our planning algorithm in simulated experiments with up to 8 sensors, and in real-world experiments using two robot arms equipped with depth cameras.

Tuan Dam, Carlo D'Eramo, Jan Peters et al.

Monte-Carlo planning and Reinforcement Learning (RL) are essential to sequential decision making. The recent AlphaGo and AlphaZero algorithms have shown how to successfully combine these two paradigms in order to solve large scale sequential decision problems. These methodologies exploit a variant of the well-known UCT algorithm to trade off exploitation of good actions and exploration of unvisited states, but their empirical success comes at the cost of poor sample-efficiency and high computation time. In this paper, we overcome these limitations by considering convex regularization in Monte-Carlo Tree Search (MCTS), which has been successfully used in RL to efficiently drive exploration. First, we introduce a unifying theory on the use of generic convex regularizers in MCTS, deriving the regret analysis and providing guarantees of exponential convergence rate. Second, we exploit our theoretical framework to introduce novel regularized backup operators for MCTS, based on the relative entropy of the policy update, and on the Tsallis entropy of the policy. Finally, we empirically evaluate the proposed operators in AlphaGo and AlphaZero on problems of increasing dimensionality and branching factor, from a toy problem to several Atari games, showing their superiority w.r.t. representative baselines.

Lauri Alho, Adrian Burian, Janne Helenius et al.

Identifying mobile network problems in 4G cells is more challenging when the complexity of the network increases, and privacy concerns limit the information content of the data. This paper proposes a data driven model for identifying 4G cells that have fundamental network throughput problems. The proposed model takes advantage of clustering and Deep Neural Networks (DNNs). Model parameters are learnt using a small number of expert-labeled data. To achieve case specific classification, we propose a model that contains a multiple clustering models block, for capturing features common for problematic cells. The captured features of this block are then used as an input to a DNN. Experiments show that the proposed model outperforms a simple classifier in identifying cells with network throughput problems. To the best of the authors' knowledge, there is no related research where network throughput classification is performed on the cell level with information gathered only from the service provider's side.