Diffusion models have been widely adopted in image and language generation and are now being applied to reinforcement learning. However, the application of diffusion models in offline cooperative Multi-Agent Reinforcement Learning (MARL) remains limited. Although existing studies explore this direction, they suffer from scalability or poor cooperation issues due to the lack of design principles for diffusion-based MARL. The Individual-Global-Max (IGM) principle is a popular design principle for cooperative MARL. By satisfying this principle, MARL algorithms achieve remarkable performance with good scalability. In this work, we extend the IGM principle to the Individual-Global-identically-Distributed (IGD) principle. This principle stipulates that the generated outcome of a multi-agent diffusion model should be identically distributed as the collective outcomes from multiple individual-agent diffusion models. We propose DoF, a diffusion factorization framework for Offline MARL. It uses noise factorization function to factorize a centralized diffusion model into multiple diffusion models. We theoretically show that the noise factorization functions satisfy the IGD principle. Furthermore, DoF uses data factorization function to model the complex relationship among data generated by multiple diffusion models. Through extensive experiments, we demonstrate the effectiveness of DoF. The source code is available at [https://github.com/xmu-rl-3dv/DoF](https://github.com/xmu-rl-3dv/DoF).

Large Language Models (LLMs) are increasingly being explored across a range of reasoning tasks. However, LLMs sometimes struggle with reasoning tasks under uncertainty that are relatively easy for humans, such as planning actions in stochastic environments. The adoption of LLMs for reasoning is impeded by uncertainty challenges, such as LLM uncertainty and environmental uncertainty. LLM uncertainty arises from the stochastic sampling process inherent to LLMs. Most LLM-based Decision-Making (LDM) approaches address LLM uncertainty through multiple reasoning chains or search trees. However, these approaches overlook environmental uncertainty, which leads to poor performance in environments with stochastic state transitions. Some recent LDM approaches deal with uncertainty by forecasting the probability of unknown variables. However, they are not designed for multi-step reasoning tasks that require interaction with the environment. To address uncertainty in LLM decision-making, we introduce PlanU, an LLM-based planning method that captures uncertainty within Monte Carlo Tree Search (MCTS). PlanU models the return of each node in the MCTS as a quantile distribution, which uses a set of quantiles to represent the return distribution. To balance exploration and exploitation during tree search, PlanU introduces an Upper Confidence Bounds with Curiosity (UCC) score which estimates the uncertainty of MCTS nodes. Through extensive experiments, we demonstrate the effectiveness of PlanU in LLM-based reasoning tasks under uncertainty.

The aeromagnetic compensation method is critical for mitigating magnetic interference in airborne geophysical surveys. In the case of complex magnetic interference, the model-driven linear regression method based on the Toles-Lawson (T-L) model cannot guarantee compensation accuracy. Additionally, purely data-driven methods often require large datasets and lack interpretability. Recent hybrid solutions, particularly Physics-Informed Neural Network (PINN), effectively combine the advantages of model-driven and data-driven methods. However, PINN is heavily reliant on manual architecture tuning, which severely limits optimization efficiency and compensation accuracy. To address this issue, this letter proposes a Differentiable Architecture Search-Guided Physics-Informed Neural Network (DARTS-PINN) method. Experimental results demonstrate that DARTS-PINN can effectively mitigate magnetic interference, improve optimization efficiency, and substantially reduce data dependence. For aeromagnetic compensation of magnetic sensors fixed inside the cabin, the root mean square error (RMSE) of DARTS-PINN is 1.089 nT. In contrast, the optimal RMSEs of the model-driven and data-driven methods are 6.195 nT and 2.026 nT, respectively.

Continuous influence maximization (CIM) in social networks aims to maximize the expected influence spreading by assigning each user a continuous weight reflecting the likelihood and cost for him becoming a seed. Traditional CIM assumes a budget constrain to limit the total cost of all users. However, this assumption does not tenable in practical applications. In practice, it is not necessary to incur costs for all the users. Instead, the budget should be set only for the cost associated with the seed set. In this article, an extended CIM problem of cost distribution under budget (CDB) is defined, which aims to assign different costs to the customers according to their ability to spread influence, ensuring that the cost of each potential seed set does not exceed the budget, while the expected spreading of the product’s influence is maximized. The NP-hardness of CDB and the monotonicity and submodularity of its objective function are investigated. We formulate the CDB problem into a constrained optimization, and present a greedy descent-based algorithm for the problem. In each iteration of the greedy descent method, the influence increment of each node is calculated according to its estimated influence spreading range. The cost distribution is updated along the direction with the maximum increment. The optimal cost distribution can be obtained after several iterations. Precision of the results obtained by the proposed algorithm is analyzed. To avoid the time-consuming simulations, we design an effective algorithm for estimating the seed influence spreading range. Experiment results on real and synthetic networks show that the proposed algorithm can significantly improve the expected influence spreading.

Multi-agent AI research promises a path to develop human-like and human-compatible intelligent technologies that complement the solipsistic view of other approaches, which mostly do not consider interactions between agents. Aiming to make progress in this direction, the Melting Pot contest 2023 focused on the problem of cooperation among interacting agents and challenged researchers to push the boundaries of multi-agent reinforcement learning (MARL) for mixed-motive games. The contest leveraged the Melting Pot environment suite to rigorously evaluate how well agents can adapt their cooperative skills to interact with novel partners in unforeseen situations. Unlike other reinforcement learning challenges, this challenge focused on social rather than environmental generalization. In particular, a population of agents performs well in Melting Pot when its component individuals are adept at finding ways to cooperate both with others in their population and with strangers. Thus Melting Pot measures cooperative intelligence.The contest attracted over 600 participants across 100+ teams globally and was a success on multiple fronts: (i) it contributed to our goal of pushing the frontiers of MARL towards building more cooperatively intelligent agents, evidenced by several submissions that outperformed established baselines; (ii) it attracted a diverse range of participants, from independent researchers to industry affiliates and academic labs, both with strong background and new interest in the area alike, broadening the field’s demographic and intellectual diversity; and (iii) analyzing the submitted agents provided important insights, highlighting areas for improvement in evaluating agents' cooperative intelligence. This paper summarizes the design aspects and results of the contest and explores the potential of Melting Pot as a benchmark for studying Cooperative AI. We further analyze the top solutions and conclude with a discussion on promising directions for future research.

We present Cycle-Contrastive Learning (CCL), a novel self-supervised method for learning video representation. Following a nature that there is a belong and inclusion relation of video and its frames, CCL is designed to find correspondences across frames and videos considering the contrastive representation in their domains respectively. It is different from recent approaches that merely learn correspondences across frames or clips. In our method, the frame and video representations are learned from a single network based on an R3D network, with a shared non-linear transformation for embedding both frame and video features before the cycle-contrastive loss. We demonstrate that the video representation learned by CCL can be transferred well to downstream tasks of video understanding, outperforming previous methods in nearest neighbour retrieval and action recognition tasks on UCF101, HMDB51 and MMAct.

Unlike vision modalities, body-worn sensors or passive sensing can avoid the failure of action understanding in vision related challenges, e.g. occlusion and appearance variation. However, a standard large-scale dataset does not exist, in which different types of modalities across vision and sensors are integrated. To address the disadvantage of vision-based modalities and push towards multi/cross modal action understanding, this paper introduces a new large-scale dataset recorded from 20 distinct subjects with seven different types of modalities: RGB videos, keypoints, acceleration, gyroscope, orientation, Wi-Fi and pressure signal. The dataset consists of more than 36k video clips for 37 action classes covering a wide range of daily life activities such as desktop-related and check-in-based ones in four different distinct scenarios. On the basis of our dataset, we propose a novel multi modality distillation model with attention mechanism to realize an adaptive knowledge transfer from sensor-based modalities to vision-based modalities. The proposed model significantly improves performance of action recognition compared to models trained with only RGB information. The experimental results confirm the effectiveness of our model on cross-subject, -view, -scene and -session evaluation criteria. We believe that this new large-scale multimodal dataset will contribute the community of multimodal based action understanding.