United States

Almost all multi-agent reinforcement learning algorithms without communication follow the principle of centralized training with decentralized execution. During the centralized training, agents can be guided by the same signals, such as the global state. However, agents lack the shared signal and choose actions given local observations during execution. Inspired by viewpoint invariance and contrastive learning, we propose consensus learning for cooperative multi-agent reinforcement learning in this study. Although based on local observations, different agents can infer the same consensus in discrete spaces without communication. We feed the inferred one-hot consensus to the network of agents as an explicit input in a decentralized way, thereby fostering their cooperative spirit. With minor model modifications, our suggested framework can be extended to a variety of multi-agent reinforcement learning algorithms. Moreover, we carry out these variants on some fully cooperative tasks and get convincing results.

AAAI 2023

Consensus Learning for Cooperative Multi-Agent Reinforcement Learning

reinforcement learning algorithms

coordination and collaboration

multiagent learning

representation learning

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-23 is the Thirty-Seventh AAAI Conference on Artificial Intelligence. The theme of this conference is to create collaborative bridges within and beyond AI. Like previous AAAI conferences, AAAI-23 will feature technical paper presentations, special tracks, invited speakers, workshops, tutorials, poster sessions, senior member presentations, competitions, and exhibit programs, and two new activities: a Bridge Program and a Lab Program. Many of these activities are tailored to the theme of bridges and all are selected according to the highest standards, with additional programs for students and young researchers. 
AAAI is providing you with a conference planner, which you can use to help organize your itinerary of activities. This includes talks to attend in person, talks to attend remotely, breaks with colleagues and your site seeing activities. To access this conference planner, please go to [https://aaai-2023.takemobi.io](https://aaai-2023.takemobi.io).

In order to access this site, you need to register. If you haven't already, please register [here](https://aaai.org/Conferences/AAAI-23/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.

poster

Abstractive summarization is the process of generating a summary given a document as input. Although significant progress has been made, the factual inconsistency between the document and the generated summary still limits its practical applications. Previous work found that the probabilities assigned by the generation model reflect its preferences for the generated summary, including the preference for factual consistency, and the preference for the language or knowledge prior as well. To separate the preference for factual consistency, we propose an unsupervised framework named CoP by controlling the preference of the generation model with the help of prompt. More specifically, the framework performs an extra inference step in which a text prompt is introduced as an additional input. In this way, another preference is described by the generation probability of this extra inference process. The difference between the above two preferences, i.e. the difference between the probabilities, could be used as measurements for detecting factual inconsistencies. Interestingly, we found that with the properly designed prompt, our framework could evaluate specific preferences and serve as measurements for fine-grained categories of inconsistency, such as entity-related inconsistency, coreference-related inconsistency, etc. Moreover, our framework could also be extended to the supervised setting to learn better prompt from the labeled data as well. Experiments show that our framework achieves new SOTA results on three factual inconsisency detection tasks.

CoP: Factual Inconsistency Detection by Controlling the Preference

Knowledge-aware question answering (KAQA) requires the model to answer questions over a knowledge base, which is essential for both open-domain QA and domain-specific QA, especially when language models alone cannot provide all the knowledge needed.
Despite the promising result of recent KAQA systems which tend to integrate linguistic knowledge from pre-trained language models (PLM) and factual knowledge from knowledge graphs (KG) to answer complex questions, a bottleneck exists in effectively fusing the representations from PLMs and KGs because of (i) the semantic and distributional gaps between them, and (ii) the difficulties in joint reasoning over the provided knowledge from both modalities.
To address the above two problems, we propose a Fine-grained Two-stage training framework (FiTs) to boost the KAQA system performance: The first stage aims at aligning representations from the PLM and the KG, thus bridging the modality gaps between them, named knowledge adaptive post-training. The second stage, called knowledge-aware fine-tuning, aims to improve the model's joint reasoning ability based on the aligned representations.
In detail,  we fine-tune the post-trained model via two auxiliary self-supervised tasks in addition to the QA supervision.
Extensive experiments demonstrate that our approach achieves state-of-the-art performance on three benchmarks in the commonsense reasoning (i.e., CommonsenseQA, OpenbookQA) and medical question answering (i.e., MedQA-USMILE) domains.

FiTs: Fine-grained Two-stage Training for Knowledge-aware Question Answering

Fine-grained urban flow inference (FUFI) problem aims at inferring the high-resolution flow maps from the coarse-grained ones, which plays an important role in sustainable and economic urban computing and traffic management. Previous models addressed the FUFI problem from spatial constraint, external factors, and memory cost. However, utilizing the new urban flow maps to calibrate the learned model is very challenging due to the ``catastrophic forgetting'' problem and is still under-explored. In this paper, we make the first step in FUFI and present CUFAR -- Continual Urban Flow inference with Adaptive knowledge Replay -- a novel framework for inferring the fine-grained citywide traffic flows. Specifically, (1) we design a spatial-temporal inference network that can extract better flow map features from both local and global levels; (2) then we present an adaptive knowledge replay (AKR) training algorithm to selectively replay the learned knowledge to facilitate the learning process of the model on new knowledge without forgetting. In addition, we also propose a knowledge discriminator to avoid ``negative replaying'' issue introduced by noisy urban flow maps. Extensive experiments on four large-scale real-world FUFI datasets demonstrate that our proposed model consistently outperforms strong baselines and effectively mitigates the forgetting problem. Source code is available at: \url{https://github.com/PattonYu/CUFAR}.

Overcoming Forgetting in Fine-Grained Urban Flow Inference via Adaptive Knowledge Replay

Unsupervised image registration commonly adopts U-Net style networks to predict dense displacement fields in the full-resolution spatial domain. For high-resolution volumetric image data, this process is however resource intensive and time-consuming. To tackle this problem, we propose the Fourier-Net, replacing the expansive path in a U-Net style network with a parameter-free model-driven decoder. Specifically, instead of our Fourier-Net learning to output a full-resolution displacement field in the spatial domain, we learn its low-dimensional representation in a band-limited Fourier domain. This representation is then decoded by our devised model-driven decoder (consisting of a zero padding layer and an inverse discrete Fourier transform layer) to the dense, full-resolution displacement field in the spatial domain. These changes allow our unsupervised Fourier-Net to contain fewer parameters and computational operations, resulting in faster inference speeds. Fourier-Net is then evaluated on two public 3D brain datasets against various state-of-the-art approaches. For example, when compared to a recent transformer-based method, i.e., TransMorph, our Fourier-Net, only using 0.22$\%$ of its parameters and 6.66$\%$ of the mult-adds, achieves a 0.6\% higher Dice score and an 11.48$\times$ faster inference speed.  Code is available at \url{https://github.com/xi-jia/Fourier-Net}.

Fourier-Net: Fast Image Registration with Band-limited Deformation

In real-world scenarios, it is crucial to build a lifelong taskoriented dialogue system (TDS) which continually adapts to new knowledge without forgetting previously acquired experiences. Existing approaches mainly focus on mitigating the catastrophic forgetting in lifelong TDS. However, the transfer ability of generalizing the accumulated old knowledge to new tasks is underexplored. In this paper, we propose a balanced meta learning and diverse sampling for lifelong task-oriented dialogue systems (called MetaLTDS), which aims to mitigate catastrophic forgetting and encourage knowledge transfer simultaneously in lifelong learning. In particular, we first overcome catastrophic forgetting by training task-specific masks to preserve the knowledge of each visited task. In addition, we employ an uncertainty-based sampling strategy to store diverse dialogue samples into the episodic memory, enhancing the sample diversity in the memory. Furthermore, we encourage knowledge transfer between tasks through balanced meta learning. We perform meta-update with a support set sampled from the current training data, and then update the model using a test set (sampled from both current task data and episodic memory). With extensive experiments on 29 tasks, we show that the proposed MetaLTDS method outperforms the strong TDS baselines in terms of both effectiveness and efficiency. For reproducibility, we submit our code anonymously at: https://github.com/anonymous2022-MetaLTDS/MetaLTDS.

Balanced Meta Learning and Diverse Sampling for Lifelong Task-oriented Dialogue Systems

When an agent votes, she typically ranks the set of available alternatives. Occasionally, she may also wish to report the intensity of her preferences by indicating adjacent pairs of alternatives in her ranking between which her preference is acutely decisive; for instance, she may suggest that she likes alternative $a$ more than $b$, but $b$ \emph{much more} than $c$. We design near-optimal voting rules which aggregate such \emph{preference rankings with intensities} using the recently-popular distortion framework. We also show that traditional voting rules, which aggregate preference rankings while ignoring (or not eliciting) intensities, can incur significant welfare loss.

Voting with Preference Intensities

Hierarchical text classification (HTC) is essential for various real applications. However, HTC models are challenging to develop because they often require processing a large volume of documents and labels with hierarchical taxonomy. Recent HTC models based on deep learning have attempted to incorporate hierarchy information into a model structure. Consequently, these models are challenging to implement when the model parameters increase for a large-scale hierarchy because the model structure depends on the hierarchy size. To solve this problem, we formulate HTC as a sub-hierarchy sequence generation to incorporate hierarchy information into a target label sequence instead of the model structure. Subsequently, we propose the Hierarchy DECoder (HiDEC), which decodes a text sequence into a sub-hierarchy sequence using recursive hierarchy decoding, classifying all parents at the same level into children at once. In addition, HiDEC is trained to use hierarchical path information from a root to each leaf in a sub-hierarchy composed of the labels of a target document via an attention mechanism and hierarchy-aware masking. HiDEC achieved state-of-the-art performance with significantly fewer model parameters than existing models on benchmark datasets, such as RCV1-v2, NYT, and EURLEX57K.

Hierarchical Text Classification As Sub-Hierarchy Sequence Generation


Powerful domain-independent planners have been developed to solve various types of planning problems. 
These planners often require a model of the acting agent's actions, given in some planning domain description language. 
Yet obtaining such an action model is a notoriously hard task. 
This task is even more challenging in mission-critical domains, where a trial-and-error approach to learning how to act is not an option. 
In such domains, the action model used to generate plans must be safe, in the sense that plans generated with it must be applicable and achieve their goals. 
The challenge of learning safe action models for planning has been recently addressed for domains in which states are sufficiently described with Boolean variables. 
In this work, we go beyond this limitation and propose the N-SAM algorithm. 
N-SAM runs in time that is polynomial in the number of observations and, under certain conditions, is guaranteed to return safe action models. 
We analyze its worst-case sample complexity, which may be intractable for benchmark domains. Empirically, however, N-SAM can quickly learn a safe action model that can solve most problems in the domain.

Learning Safe Numeric Action Models

Autoregressive language modeling (ALM) have been successfully used in self-supervised pre-training in Natural language processing (NLP). However, this paradigm has not achieved comparable results with other self-supervised approach in computer vision (e.g., contrastive learning, mask image modeling). In this paper, we try to find the reason why autoregressive modeling does not work well on vision tasks. To tackle this problem, we fully analyze the limitation of visual autoregressive methods and proposed a novel stochastic autoregressive image modeling (named SAIM) by the two simple designs. First, we employ stochastic permutation strategy to generate effective and robust image context which is critical for vision tasks. Second, we create a parallel encoder-decoder training process in which the encoder serves a similar role to the standard vision transformer focus on learning the whole contextual information, and meanwhile the decoder predicts the content of the current position, so that the encoder and decoder can reinforce each other. By introducing stochastic prediction and the parallel encoder-decoder, SAIM significantly improve the performance of autoregressive image modeling. Our method achieves the best accuracy (83.9\%) on the vanilla ViT-Base model among methods using only ImageNet-1K data. Transfer performance in downstream tasks also show that our model achieves competitive performance.

Exploring Stochastic Autoregressive Image Modeling for Visual Representation

Nowadays, systems based on machine learning (ML) are widely used in different domains.
Given their popularity, ML models have become targets for various attacks.
As a result, research at the intersection of security/privacy and ML has flourished. Typically such work has focused on individual types of security/privacy concerns and mitigations thereof.

However, in real-life deployments, an ML model will need to be protected against several concerns simultaneously.
A protection mechanism optimal for a specific security or privacy concern may interact negatively with mechanisms intended to address other concerns. Despite its practical relevance, the potential for such conflicts has not been studied adequately.

In this work, we first provide a framework for analyzing such conflicting interactions.
We then focus on systematically analyzing pairwise interactions between protection mechanisms for one concern, model and data ownership verification, with two other classes of ML protection mechanisms: differentially private training, and robustness against model evasion.
We find that several pairwise interactions result in conflicts.

We also explore potential approaches for avoiding such conflicts. First, we study the effect of hyperparameter relaxations, finding that there is no sweet spot balancing the performance of both protection mechanisms.
Second, we explore whether modifying one type of protection mechanism (ownership verification) so as to decouple it from factors that may be impacted by a conflicting mechanism (differentially private training or robustness to model evasion) can avoid conflict.
We show that this approach can indeed avoid the conflict between ownership verification mechanisms when combined with differentially private training, but has no effect on robustness to model evasion. We conclude by identifying the gaps in the landscape of studying interactions between other types of ML protection mechanisms.

Premium content

Consensus Learning for Cooperative Multi-Agent Reinforcement Learning

Downloads

Next from AAAI 2023

CoP: Factual Inconsistency Detection by Controlling the Preference

Similar lecture

HAVEN: Hierarchical Cooperative Multi-Agent Reinforcement Learning with Dual Coordination Mechanism