Canada

A fundamental challenge in multi-agent reinforcement learning (MARL) is to learn the joint policy in an extremely large search space, which grows exponentially with the number of agents. Moreover, fully decentralized policy factorization significantly restricts the search space, which may lead to sub-optimal policies. In contrast, the auto-regressive joint policy can represent a much richer class of joint policies by factorizing the joint policy into the product of a series of conditional individual policies. While such factorization introduces the action dependency among agents explicitly in sequential execution, it does not take full advantage of the dependency during learning. In particular, the subsequent agents do not give the preceding agents feedback about their decisions. In this paper, we propose a new framework Back-Propagation Through Agents (BPTA) that directly accounts for both agents&#39; own policy updates and the learning of their dependent counterparts. This is achieved by propagating the feedback through action chains. With the proposed framework, our Bidirectional Proximal Policy Optimisation (BPPO) outperforms the state-of-the-art methods. Extensive experiments on matrix games, StarCraftII v2, Multi-agent MuJoCo, and Google Research Football demonstrate the effectiveness of the proposed method.

AAAI 2024

Backpropagation Through Agents

reinforcement learning algorithms

multiagent learning

We are pleased to announce the Thirty-Ninth AAAI Conference on Artificial Intelligence (AAAI-25), which will be held in Philadelphia, Pennsylvania at the Pennsylvania Convention Center from February 25 to March 4, 2025.

The purpose of the AAAI conference series is to promote research in Artificial Intelligence (AI) and foster scientific exchange between researchers, practitioners, scientists, students, and engineers across the entirety of AI and its affiliated disciplines. AAAI-25 will feature technical paper presentations, special tracks, invited speakers, workshops, tutorials, poster sessions, senior member presentations, competitions, and exhibit programs, and a range of other activities to be announced.

### [Invited Speakers](https://aaai.org/conference/aaai/aaai-25/aaai-25-invited-speakers/)

Register [here](https://aaai.org/conference/aaai/aaai-25/registration/)

AAAI 2025

The purpose of the AAAI conference series is to promote research in Artificial Intelligence (AI) and foster scientific exchange between researchers, practitioners, scientists, students, and engineers across the entirety of AI and its affiliated disciplines. AAAI-25 will feature technical paper presentations, special tracks, invited speakers, workshops, tutorials, poster sessions, senior member presentations, competitions, and exhibit programs, and a range of other activities to be announced.



A fundamental challenge in multi-agent reinforcement learning (MARL) is to learn the joint policy in an extremely large search space, which grows exponentially with the number of agents. Moreover, fully decentralized policy factorization significantly restricts the search space, which may lead to sub-optimal policies. In contrast, the auto-regressive joint policy can represent a much richer class of joint policies by factorizing the joint policy into the product of a series of conditional individual policies. While such factorization introduces the action dependency among agents explicitly in sequential execution, it does not take full advantage of the dependency during learning. In particular, the subsequent agents do not give the preceding agents feedback about their decisions. In this paper, we propose a new framework Back-Propagation Through Agents (BPTA) that directly accounts for both agents' own policy updates and the learning of their dependent counterparts. This is achieved by propagating the feedback through action chains. With the proposed framework, our Bidirectional Proximal Policy Optimisation (BPPO) outperforms the state-of-the-art methods. Extensive experiments on matrix games, StarCraftII v2, Multi-agent MuJoCo, and Google Research Football demonstrate the effectiveness of the proposed method.

technical paper

We are pleased to announce the Thirty-Eighth AAAI Conference on Artificial Intelligence (AAAI-24), which will be held in Vancouver, British Columbia at the Vancouver Convention Centre – West Building from 20-27 February, 2024.

The purpose of the AAAI conference series is to promote research in Artificial Intelligence (AI) and foster scientific exchange between researchers, practitioners, scientists, students, and engineers across the entirety of AI and its affiliated disciplines. AAAI-24 will feature technical paper presentations, special tracks, invited speakers, workshops, tutorials, poster sessions, senior member presentations, competitions, and exhibit programs, and a range of other activities to be announced.

We expect for AAAI-24 to be an in-person conference – one author of all accepted papers will be expected to present work in person unless there are exceptional circumstances that prevent this.

In order to access the AAAI-24 event page you need to register [here](https://aaai.org/aaai-conference/registration/)

The purpose of the AAAI conference series is to promote research in Artificial Intelligence (AI) and foster scientific exchange between researchers, practitioners, scientists, students, and engineers across the entirety of AI and its affiliated disciplines. 

Multi-goal conversational recommender system (MG-CRS) which is more in line with realistic scenarios has attracted a lot of attention. MG-CRS can dynamically capture the demands of users in conversation, continuously engage their interests, and make recommendations. The key of accomplishing these tasks is to plan a reasonable goal sequence which can naturally guide the user to accept the recommended goal. Previous works have demonstrated that mining the correlations of goals from the goal sequences in the dialogue corpus is helpful for recommending the goal that the user is interested in. However, they independently model correlations for each level of goal (i.e., goal type or entity) and neglect the order of goals appear in the dialogue. In this paper, we propose a goal interaction graph planning framework which constructs a directed heterogeneous graph to flexibly model the correlations between any level of goals and retain the order of goals. We design a goal interaction graph learning module to model the goal correlations and propagate goal representations via directed edges, then use an encoder and a dual-way fusion decoder to extract the most relevant information with the current goal from the conversation and domain knowledge, making the next-goal prediction fully exploit the prior goal correlations and user feedback. Finally we generate engaging responses based on the predicted goal sequence to complete the recommendation task. Experiments on two benchmark datasets show that our method achieves significant improvements in both the goal planning and response generation tasks.

A Goal Interaction Graph Planning Framework for Conversational Recommendation

Human brains can continually acquire and learn new skills and knowledge over time from a dynamically changing environment without forgetting previously learnt information. Such a capacity can selectively transfer some important and recently seen information to the persistent knowledge regions of the brain. Inspired by this intuition, we propose a new memory-based approach for image reconstruction and generation in continual learning, consisting of a temporary and evolving memory, with two different storage strategies, corresponding to the temporary and permanent memorisation. The temporary memory aims to preserve up-to-date information while the evolving memory can dynamically increase its capacity in order to preserve permanent knowledge information. This is achieved by the proposed memory expansion mechanism that selectively transfers those data samples deemed as important from the temporary memory to new clusters defined within the evolved memory according to an information novelty criterion. Such a mechanism promotes the knowledge diversity among clusters in the evolved memory, resulting in capturing more diverse information by using a compact memory capacity. Furthermore, we propose a two-step optimization strategy for training a Variational Autoencoder (VAE) to implement generation and representation learning tasks, which updates the generator and inference models separately using two optimisation paths. This approach leads to a better trade-off between generation and reconstruction performance. We show empirically and theoretically that the proposed approach can learn meaningful latent representations while generating diverse images from different domains. The source code and supplementary material (SM) are available at https://github.com/dtuzi123/DEMC.

Task-Free Continual Generation and Representation Learning via Dynamic Expansionable Memory Cluster

Video recognition systems are vulnerable to adversarial examples. Recent studies show that style transfer-based and patch-based unrestricted perturbations can effectively improve attack efficiency. These attacks, however, face two main challenges: 1) Adding large stylized perturbations to all pixels reduces the naturalness of the video and such perturbations can be easily detected. 2) Patch-based video attacks are not extensible to targeted attacks due to the limited search space of reinforcement learning that has been widely used in video attacks recently. In this paper, we focus on the video black-box setting and propose a novel attack framework named LogoStyleFool by adding a stylized logo to the clean video. We separate the attack into three stages: style reference selection, reinforcement-learning-based logo style transfer, and perturbation optimization. We solve the first challenge by scaling down the perturbation range to a regional logo, while the second challenge is addressed by complementing an optimization stage after reinforcement learning. Experimental results substantiate the overall superiority of LogoStyleFool over three state-of-the-art patch-based attacks in terms of attack performance and semantic preservation. Meanwhile, LogoStyleFool still maintains its performance against two existing patch-based defense methods. We believe that our research is beneficial in increasing the attention of the security community to such subregional style transfer attacks.

LogoStyleFool: Vitiating Video Recognition Systems via Logo Style Transfer

Image harmonization aims to generate a more realistic appearance of foreground and background for a composite image. All the existing methods perform the same harmonization process for the whole foreground. However, the implanted foreground always contains different appearance patterns. Existing solutions ignore the difference of each color block and lose some specific details. Therefore, we propose a novel global-local two stages framework for Fine-grained Region-aware Image Harmonization (FRIH). In the first stage, the whole input foreground mask is used to make a global coarse-grained harmonization. In the second stage, we adaptively cluster the input foreground mask into several submasks. Each submask and the coarsely adjusted image are concatenated respectively and fed into a lightweight cascaded module, refining the global harmonization result. Moreover, we further design a fusion prediction module to generate the final result, utilizing the different degrees of harmonization results comprehensively. Without bells and whistles, our FRIH achieves a competitive performance on iHarmony4 dataset with a lightweight model.

FRIH: Fine-Grained Region-Aware Image Harmonization | VIDEO

Recently, several lightweight methods have been proposed to implement single-image super-resolution (SISR) on resource-constrained devices. However, these methods primarily focus on simplifying network structures without the full utilization of shallow features. The fact remains that shallow features encompass crucial details for the super-resolution task, including edges, textures, and colors. Therefore, developing a novel architecture that can effectively integrate features from different levels and capitalize on their mutual complementarity is necessary. We first analyze the relationship between multi-stage features and the restoration tasks in a classic lightweight SR method. Based on these observations, we propose an Omni-Stage Feature Fusion (OSFF) architecture, which incorporates Original Image Stacked Initialisation, Shallow Feature Global Connection, and Multi-Receptive Field Dynamic Fusion. An Attention-Enhanced Feature Distillation module is also designed to enhance the model performance. Finally, leveraging these contributions, we construct an Omni-Stage Feature Fusion Network (OSFFNet). Through extensive experiments on various benchmark datasets, the proposed model outperforms state-of-the-art methods. Notably, it achieves a 0.26dB PSNR improvement over the second-best method for x2 SR on the Urban100 dataset.

OSFFNet: Omni-Stage Feature Fusion Network for Lightweight Image Super-resolution

Envy-freeness is one of the most important fairness concerns when allocating items. We study envy-free house allocation when agents have uncertain preferences over items and consider several well-studied preference uncertainty models. The central problem that we focus on is computing an allocation that has the highest probability of being envy-free. We show that each model leads to a distinct set of algorithmic and complexity results, including detailed results on (in-)approximability. En route, we consider two related problems of checking whether there exists an allocation that is possibly or necessarily envy-free. We give a complete picture of the computational complexity of these two problems for all the uncertainty models we consider.

Envy-Free House Allocation under Uncertain Preferences

Lowering the memory requirement in full-parameter training on large models has become a hot research area. MeZO fine-tunes the large language models (LLMs) by just forward passes in a zeroth-order SGD optimizer (ZO-SGD), demonstrating excellent performance with the same GPU memory usage as inference. However, the simulated perturbation stochastic approximation for gradient estimate in MeZO leads to severe oscillations and incurs a substantial time overhead. Moreover, without momentum regularization, MeZO shows severe over-fitting problems. Lastly, the perturbation-irrelevant momentum on ZO-SGD does not improve the convergence rate. This study proposes ZO-AdaMU to resolve the above problems by adapting the simulated perturbation with momentum in its stochastic approximation. Unlike existing adaptive momentum methods, we relocate momentum on simulated perturbation in stochastic gradient approximation. Our convergence analysis and experiments prove this is a better way to improve convergence stability and rate in ZO-SGD. Extensive experiments demonstrate that ZO-AdaMU yields better generalization for LLMs fine-tuning across various NLP tasks than MeZO and its momentum variants.

ZO-AdaMU Optimizer: Adapting Perturbation by the Momentum and Uncertainty in Zeroth-Order Optimization

The success of Graph Neural Networks (GNNs) in practice has motivated extensive research on their theoretical properties. This includes recent results that characterise node classifiers expressible by GNNs in terms of first order logic. Most of the analysis, however, has been focused on GNNs with fixed number of message-passing iterations (i.e., layers), which cannot realise many simple classifiers such as reachability of a node with a given label. In this paper, we start to fill this gap and study the foundations of GNNs that can perform more than a fixed number of message-passing iterations. We first formalise two generalisations of the basic GNNs: recurrent GNNs (RecGNNs), which repeatedly apply message-passing iterations until the node classifications become stable, and graph-size GNNs (GSGNNs), which exploit a built-in function of the input graph size to decide the number of message-passings. We then formally prove that GNN classifiers are strictly less expressive than RecGNN ones, and RecGNN classifiers are strictly less expressive than GSGNN ones. To get this result, we identify novel semantic characterisations of the three formalisms in terms of suitable variants of bisimulation, which we believe have their own value for our understanding of GNNs. Finally, we prove syntactic logical characterisations of RecGNNs and GSGNNs analogous to the logical characterisation of plain GNNs, where we connect the two formalisms to monadic monotone fixpoint logic---a generalisation of first-order logic that supports recursion.

Recurrent Graph Neural Networks and Their Connections to Bisimulation and Logic

Detecting OOD inputs is crucial to deploy machine learning models to the real world safely. However, existing OOD detection methods require an in-distribution (ID) dataset to retrain the models. In this paper, we propose a Deep Generative Models (DGMs) based transferable OOD detection that does not require retraining on the new ID dataset. We first establish and substantiate two hypotheses on DGMs: DGMs exhibit a predisposition towards acquiring low-level features, in preference to semantic information; the lower bound of DGM's log-likelihoods is tied to the conditional entropy between the model input and target output. Drawing on the aforementioned hypotheses, we present an innovative image-erasing strategy, which is designed to create distinct conditional entropy distributions for each individual ID dataset. By training a DGM on a complex dataset with the proposed image-erasing strategy, the DGM could capture the discrepancy of conditional entropy distribution for varying ID datasets, without re-training. We validate the proposed method on the five datasets and show that, without retraining, our method achieves comparable performance to the state-of-the-art group-based OOD detection methods. The project codes will be open-sourced on our project website.

Learning by Erasing: Conditional Entropy Based Transferable Out-of-Distribution Detection | VIDEO

Estimating 3D full-body pose from sparse sensor data is a pivotal technique employed for the reconstruction of realistic human motions in Augmented Reality and Virtual Reality. However, translating sparse sensor signals into comprehensive human motion remains a challenge since the sparsely distributed sensors in common VR systems fail to capture the motion of full human body. In this paper, we use well-designed Body Pose Graph (BPG) to represent the human body and translate the challenge into a prediction problem of graph missing nodes. Then, we propose a novel full-body motion reconstruction framework based on BPG. To establish BPG, nodes are initially endowed with features extracted from sparse sensor signals. Features from identifiable joint nodes across diverse sensors are amalgamated and processed from both temporal and spatial perspectives. Temporal dynamics are captured using the Temporal Pyramid Structure, while spatial relations in joint movements inform the spatial attributes. The resultant features serve as the foundational elements of the BPG nodes. To further refine the BPG, node features are updated through a graph neural network that incorporates edge reflecting varying joint relations. Our method's effectiveness is evidenced by the attained state-of-the-art performance, particularly in lower body motion, outperforming other baseline methods. Additionally, an ablation study validates the efficacy of each module in our proposed framework.

Downloads

Next from AAAI 2024

A Goal Interaction Graph Planning Framework for Conversational Recommendation

Similar lecture

Cooperative Policy Agreement: Learning Diverse Policy for Offline MARL