Autonomous robot person-following (RPF) systems are crucial for personal assistance and security but suffer from target loss due to occlusions in dynamic, unknown environments. Current methods rely on pre-built maps and assume static environments, limiting their effectiveness in real-world settings. There is a critical gap in re-finding targets under topographic (e.g., walls, corners) and dynamic (e.g., moving pedestrians) occlusions. In this paper, we propose a novel heuristic-guided search framework that dynamically builds environmental maps while following the target and resolves various occlusions by prioritizing high-probability areas for locating the target. For topographic occlusions, a belief-guided search field is constructed and used to evaluate the likelihood of the target's presence, while for dynamic occlusions, a fluid-field approach allows the robot to adaptively follow or overtake moving occluders. Past motion cues and environmental observations refine the search decision over time. Our results demonstrate that the proposed method outperforms existing approaches in terms of search efficiency and success rates, both in simulations and real-world tests. Our target search method enhances the adaptability and reliability of RPF systems in unknown and dynamic environments to support their use in real-world applications. Our code, video, experimental results and appendix are available at https://medlartea.github.io/rpf-search/.

OV AMOS: A Framework for Open-V ocabulary Multi-Object Search in Unknown Environments Qianwei Wang*, Yifan Xu*, Vineet Kamat, and Carol Menassa Abstract -- Object search is a fundamental task for robots deployed in indoor building environments, yet challenges arise due to observation instability, especially for open-vocabulary models. While foundation models (LLMs/VLMs) enable reasoning about object locations even without direct visibility, the ability to recover from failures and replan remains crucial. T o address these challenges, we propose a framework integrating VLM-based reasoning, frontier-based exploration, and a Partially Observable Markov Decision Process (POMDP) framework to solve the MOS problem in novel environments. VLM enhances search efficiency by inferring object-environment relationships, frontier-based exploration guides navigation in unknown spaces, and POMDP models observation uncertainty, allowing recovery from failures in occlusion and cluttered environments. We evaluate our framework on 120 simulated scenarios across several Habitat-Matterport3D (HM3D) scenes and a real-world robot experiment in a 50-square-meter office, demonstrating significant improvements in both efficiency and success rate over baseline methods. I NTRODUCTION Multi-Object Search (MOS) is a crucial task in robotics [1]. Consider a scenario where in a workplace setting, a robot may need to retrieve multiple objects to complete a task, such as gathering necessary documents, tools, or equipment for an assembly process.

In the context of multi-agent reinforcement learning, generalization is a challenge to solve various tasks that may require different joint policies or coordination without relying on policies specialized for each task. We refer to this type of problem as a multi-task, and we train agents to be versatile in this multi-task setting through a single training process. To address this challenge, we introduce TRajectory-class-Aware Multi-Agent reinforcement learning (TRAMA). In TRAMA, agents recognize a task type by identifying the class of trajectories they are experiencing through partial observations, and the agents use this trajectory awareness or prediction as additional information for action policy. To this end, we introduce three primary objectives in TRAMA: (a) constructing a quantized latent space to generate trajectory embeddings that reflect key similarities among them; (b) conducting trajectory clustering using these trajectory embeddings; and (c) building a trajectory-class-aware policy. Specifically for (c), we introduce a trajectory-class predictor that performs agent-wise predictions on the trajectory class; and we design a trajectory-class representation model for each trajectory class. Each agent takes actions based on this trajectory-class representation along with its partial observation for task-aware execution. The proposed method is evaluated on various tasks, including multi-task problems built upon StarCraft II. Empirical results show further performance improvements over state-of-the-art baselines.

We study the problem of estimating Dynamic Discrete Choice (DDC) models, also known as offline Maximum Entropy-Regularized Inverse Reinforcement Learning (offline MaxEnt-IRL) in machine learning. The objective is to recover reward or $Q^*$ functions that govern agent behavior from offline behavior data. In this paper, we propose a globally convergent gradient-based method for solving these problems without the restrictive assumption of linearly parameterized rewards. The novelty of our approach lies in introducing the Empirical Risk Minimization (ERM) based IRL/DDC framework, which circumvents the need for explicit state transition probability estimation in the Bellman equation. Furthermore, our method is compatible with non-parametric estimation techniques such as neural networks. Therefore, the proposed method has the potential to be scaled to high-dimensional, infinite state spaces. A key theoretical insight underlying our approach is that the Bellman residual satisfies the Polyak-Lojasiewicz (PL) condition -- a property that, while weaker than strong convexity, is sufficient to ensure fast global convergence guarantees. Through a series of synthetic experiments, we demonstrate that our approach consistently outperforms benchmark methods and state-of-the-art alternatives.

This paper proposes an online inference method of the stochastic gradient descent (SGD) with a constant learning rate for quantile loss functions with theoretical guarantees. Since the quantile loss function is neither smooth nor strongly convex, we view such SGD iterates as an irreducible and positive recurrent Markov chain. By leveraging this interpretation, we show the existence of a unique asymptotic stationary distribution, regardless of the arbitrarily fixed initialization. To characterize the exact form of this limiting distribution, we derive bounds for its moment generating function and tail probabilities, controlling over the first and second moments of SGD iterates. By these techniques, we prove that the stationary distribution converges to a Gaussian distribution as the constant learning rate $\eta\rightarrow0$. Our findings provide the first central limit theorem (CLT)-type theoretical guarantees for the last iterate of constant learning-rate SGD in non-smooth and non-strongly convex settings. We further propose a recursive algorithm to construct confidence intervals of SGD iterates in an online manner. Numerical studies demonstrate strong finite-sample performance of our proposed quantile estimator and inference method. The theoretical tools in this study are of independent interest to investigate general transition kernels in Markov chains.

We investigate off-policy evaluation (OPE), a central and fundamental problem in reinforcement learning (RL), in the challenging setting of Partially Observable Markov Decision Processes (POMDPs) with large observation spaces. Recent works of Uehara et al. (2023a); Zhang & Jiang (2024) developed a model-free framework and identified important coverage assumptions (called belief and outcome coverage) that enable accurate OPE of memoryless policies with polynomial sample complexities, but handling more general target policies that depend on the entire observable history remained an open problem. In this work, we prove information-theoretic hardness for model-free OPE of history-dependent policies in several settings, characterized by additional assumptions imposed on the behavior policy (memoryless vs. history-dependent) and/or the state-revealing property of the POMDP (single-step vs. multi-step revealing). We further show that some hardness can be circumvented by a natural model-based algorithm -- whose analysis has surprisingly eluded the literature despite the algorithm's simplicity -- demonstrating provable separation between model-free and model-based OPE in POMDPs.

-- Recent advances in large language models (LLMs) have demonstrated their potential as planners in human-robot collaboration (HRC) scenarios, offering a promising alternative to traditional planning methods. LLMs, which can generate structured plans by reasoning over natural language inputs, have the ability to generalize across diverse tasks and adapt to human instructions. This paper investigates the potential of LLMs to facilitate planning in the context of human-robot collaborative tasks, with a focus on their ability to reason from high-level, vague human inputs, and fine-tune plans based on real-time feedback. We propose a novel hybrid framework that combines LLMs with human feedback to create dynamic, context-aware task plans. Our work also highlights how a single, concise prompt can be used for a wide range of tasks and environments, overcoming the limitations of long, detailed structured prompts typically used in prior studies. By integrating user preferences into the planning loop, we ensure that the generated plans are not only effective but aligned with human intentions. Planning is a fundamental aspect of robotics, enabling autonomous agents to generate sequences of actions to achieve specific goals. Traditional planning methods in the context of human-robot collaboration (HRC) and assistive robots can be broadly categorized into two main types: rule-based planners and learning-based planners . Rule-based planners rely on predefined heuristics and symbolic representations, making them interpretable at the expense of not being able to adapt to complex or dynamic environments. In contrast, learning-based planners, particularly those utilizing deep reinforcement learning, learn to generate plans from experience in an adaptive manner.

Markov decision processes (MDPs) are a canonical model to reason about decision making within a stochastic environment. We study a fundamental class of infinite MDPs: one-counter MDPs (OC-MDPs). They extend finite MDPs via an associated counter taking natural values, thus inducing an infinite MDP over the set of configurations (current state and counter value). We consider two characteristic objectives: reaching a target state (state-reachability), and reaching a target state with counter value zero (selective termination). The synthesis problem for the latter is not known to be decidable and connected to major open problems in number theory. Furthermore, even seemingly simple strategies (e.g., memoryless ones) in OC-MDPs might be impossible to build in practice (due to the underlying infinite configuration space): we need finite, and preferably small, representations. To overcome these obstacles, we introduce two natural classes of concisely represented strategies based on a (possibly infinite) partition of counter values in intervals. For both classes, and both objectives, we study the verification problem (does a given strategy ensure a high enough probability for the objective?), and two synthesis problems (does there exist such a strategy?): one where the interval partition is fixed as input, and one where it is only parameterized. We develop a generic approach based on a compression of the induced infinite MDP that yields decidability in all cases, with all complexities within PSPACE.

The AAAI 2025 outstanding paper awards were announced during the opening ceremony of the 39th Annual AAAI Conference on Artificial Intelligence on Thursday 27 February. Papers are recommended for consideration during the review process by members of the Program Committee. This year, three papers have been selected as outstanding papers, with a further paper being recognised in the special track on AI for social impact. Abstract: A fundamental task in multi-agent systems is to match agents to alternatives (e.g., resources or tasks). Often, this is accomplished by eliciting agents' ordinal rankings over the alternatives instead of their exact numerical utilities.

In this paper, we explore the fundamental role of the Monge-Amp\`ere equation in deep learning, particularly in the context of Boltzmann machines and energy-based models. We first review the structure of Boltzmann learning and its relation to free energy minimization. We then establish a connection between optimal transport theory and deep learning, demonstrating how the Monge-Amp\`ere equation governs probability transformations in generative models. Additionally, we provide insights from quantum geometry, showing that the space of covariance matrices arising in the learning process coincides with the Connes-Araki-Haagerup (CAH) cone in von Neumann algebra theory. Furthermore, we introduce an alternative approach based on renormalization group (RG) flow, which, while distinct from the optimal transport perspective, reveals another manifestation of the Monge-Amp\`ere domain in learning dynamics. This dual perspective offers a deeper mathematical understanding of hierarchical feature learning, bridging concepts from statistical mechanics, quantum geometry, and deep learning theory.