We study the allocation of shared resources over multiple rounds among competing agents, via a dynamic max-min fair (DMMF) mechanism: the good in each round is allocated to the requesting agent with the least number of allocations received to date. Previous work has shown that when an agent has i.i.d. values across rounds, then in the worst case, she can never get more than a constant strictly less than $1$ fraction of her ideal utility -- her highest achievable utility given her nominal share of resources. Moreover, an agent can achieve at least half her utility under carefully designed `pseudo-market' mechanisms, even though other agents may act in an arbitrary (possibly adversarial and collusive) manner. We show that this robustness guarantee also holds under the much simpler DMMF mechanism. More significantly, under mild assumptions on the value distribution, we show that DMMF in fact allows each agent to realize a $1 - o(1)$ fraction of her ideal utility, despite arbitrary behavior by other agents. We achieve this by characterizing the utility achieved under a richer space of strategies, wherein an agent can tune how aggressive to be in requesting the item. Our new strategies also allow us to handle settings where an agent's values are correlated across rounds, thereby allowing an adversary to predict and block her future values. We prove that again by tuning one's aggressiveness, an agent can guarantee $\Omega(\gamma)$ fraction of her ideal utility, where $\gamma\in [0, 1]$ is a parameter that quantifies dependence across rounds (with $\gamma = 1$ indicating full independence and lower values indicating more correlation). Finally, we extend our efficiency results to the case of reusable resources, where an agent might need to hold the item over multiple rounds to receive utility.
相關內容
Large Language Models exhibit robust problem-solving capabilities for diverse tasks. However, most LLM-based agents are designed as specific task solvers with sophisticated prompt engineering, rather than agents capable of learning and evolving through interactions. These task solvers necessitate manually crafted prompts to inform task rules and regulate LLM behaviors, inherently incapacitating to address complex dynamic scenarios e.g., large interactive games. In light of this, we propose Agent-Pro: an LLM-based Agent with Policy-level Reflection and Optimization that can learn a wealth of expertise from interactive experiences and progressively elevate its behavioral policy. Specifically, it involves a dynamic belief generation and reflection process for policy evolution. Rather than action-level reflection, Agent-Pro iteratively reflects on past trajectories and beliefs, fine-tuning its irrational beliefs for a better policy. Moreover, a depth-first search is employed for policy optimization, ensuring continual enhancement in policy payoffs. Agent-Pro is evaluated across two games: Blackjack and Texas Hold'em, outperforming vanilla LLM and specialized models. Our results show Agent-Pro can learn and evolve in complex and dynamic scenes, which also benefits numerous LLM-based applications.
Addressing the challenges related to data sparsity, cold-start problems, and diversity in recommendation systems is both crucial and demanding. Many current solutions leverage knowledge graphs to tackle these issues by combining both item-based and user-item collaborative signals. A common trend in these approaches focuses on improving ranking performance at the cost of escalating model complexity, reducing diversity, and complicating the task. It is essential to provide recommendations that are both personalized and diverse, rather than solely relying on achieving high rank-based performance, such as Click-through Rate, Recall, etc. In this paper, we propose a hybrid multi-task learning approach, training on user-item and item-item interactions. We apply item-based contrastive learning on descriptive text, sampling positive and negative pairs based on item metadata. Our approach allows the model to better understand the relationships between entities within the knowledge graph by utilizing semantic information from text. It leads to more accurate, relevant, and diverse user recommendations and a benefit that extends even to cold-start users who have few interactions with items. We perform extensive experiments on two widely used datasets to validate the effectiveness of our approach. Our findings demonstrate that jointly training user-item interactions and item-based signals using synopsis text is highly effective. Furthermore, our results provide evidence that item-based contrastive learning enhances the quality of entity embeddings, as indicated by metrics such as uniformity and alignment.
This research addresses the critical necessity for advanced rapid response operations in managing a spectrum of environmental hazards. We propose a novel framework, qIoV that integrates quantum computing with the Internet-of-Vehicles (IoV) to leverage the computational efficiency, parallelism, and entanglement properties of quantum mechanics. Our approach involves the use of environmental sensors mounted on vehicles for precise air quality assessment. These sensors are designed to be highly sensitive and accurate, leveraging the principles of quantum mechanics to detect and measure environmental parameters. A salient feature of our proposal is the Quantum Mesh Network Fabric (QMF), a system designed to dynamically adjust the quantum network topology in accordance with vehicular movements. This capability is critical to maintaining the integrity of quantum states against environmental and vehicular disturbances, thereby ensuring reliable data transmission and processing. Moreover, our methodology is further augmented by the incorporation of a variational quantum classifier (VQC) with advanced quantum entanglement techniques. This integration offers a significant reduction in latency for hazard alert transmission, thus enabling expedited communication of crucial data to emergency response teams and the public. Our study on the IBM OpenQSAM 3 platform, utilizing a 127 Qubit system, revealed significant advancements in pair plot analysis, achieving over 90% in precision, recall, and F1-Score metrics and an 83% increase in the speed of toxic gas detection compared to conventional methods.Additionally, theoretical analyses validate the efficiency of quantum rotation, teleportation protocols, and the fidelity of quantum entanglement, further underscoring the potential of quantum computing in enhancing analytical performance.
We study networks of processes that all execute the same finite protocol and communicate synchronously in two different ways: a process can broadcast one message to all other processes or send it to at most one other process. In both cases, if no process can receive the message, it will still be sent. We establish a precise complexity class for two coverability problems with a parameterised number of processes: the state coverability problem and the configuration coverability problem. It is already known that these problems are Ackermann-hard (but decidable) in the general case. We show that when the protocol is Wait-Only, i.e., it has no state from which a process can send and receive messages, the complexity drops to P and PSPACE, respectively.
We study a family of distance functions on rankings that allow for asymmetric treatments of alternatives and consider the distinct relevance of the top and bottom positions for ordered lists. We provide a full axiomatic characterization of our distance. In doing so, we retrieve new characterizations of existing axioms and show how to effectively weaken them for our purposes. This analysis highlights the generality of our distance as it embeds many (semi)metrics previously proposed in the literature. Subsequently, we show that, notwithstanding its level of generality, our distance is still readily applicable. We apply it to preference aggregation, studying the features of the associated median voting rule. It is shown how the derived preference function satisfies many desirable features in the context of voting rules, ranging from fairness to majority and Pareto-related properties. We show how to compute consensus rankings exactly, and provide generalized Diaconis-Graham inequalities that can be leveraged to obtain approximation algorithms. Finally, we propose some truncation ideas for our distances inspired by Lu and Boutilier (2010). These can be leveraged to devise a Polynomial-Time-Approximation Scheme for the corresponding rank aggregation problem.
The persistent issue of wrongful convictions in the United States emphasizes the need for scrutiny and improvement of the criminal justice system. While statistical methods for the evaluation of forensic evidence, including glass, fingerprints, and DNA, have significantly contributed to solving intricate crimes, there is a notable lack of national-level standards to ensure the appropriate application of statistics in forensic investigations. We discuss the obstacles in the application of statistics in court, and emphasize the importance of making statistical interpretation accessible to non-statisticians, especially those who make decisions about potentially innocent individuals. We investigate the use and misuse of statistical methods in crime investigations, in particular the likelihood ratio approach. We further describe the use of graphical models, where hypotheses and evidence can be represented as nodes connected by arrows signifying association or causality. We emphasize the advantages of special graph structures, such as object-oriented Bayesian networks and chain event graphs, which allow for the concurrent examination of evidence of various nature.
Multimodality Representation Learning, as a technique of learning to embed information from different modalities and their correlations, has achieved remarkable success on a variety of applications, such as Visual Question Answering (VQA), Natural Language for Visual Reasoning (NLVR), and Vision Language Retrieval (VLR). Among these applications, cross-modal interaction and complementary information from different modalities are crucial for advanced models to perform any multimodal task, e.g., understand, recognize, retrieve, or generate optimally. Researchers have proposed diverse methods to address these tasks. The different variants of transformer-based architectures performed extraordinarily on multiple modalities. This survey presents the comprehensive literature on the evolution and enhancement of deep learning multimodal architectures to deal with textual, visual and audio features for diverse cross-modal and modern multimodal tasks. This study summarizes the (i) recent task-specific deep learning methodologies, (ii) the pretraining types and multimodal pretraining objectives, (iii) from state-of-the-art pretrained multimodal approaches to unifying architectures, and (iv) multimodal task categories and possible future improvements that can be devised for better multimodal learning. Moreover, we prepare a dataset section for new researchers that covers most of the benchmarks for pretraining and finetuning. Finally, major challenges, gaps, and potential research topics are explored. A constantly-updated paperlist related to our survey is maintained at //github.com/marslanm/multimodality-representation-learning.
A fundamental goal of scientific research is to learn about causal relationships. However, despite its critical role in the life and social sciences, causality has not had the same importance in Natural Language Processing (NLP), which has traditionally placed more emphasis on predictive tasks. This distinction is beginning to fade, with an emerging area of interdisciplinary research at the convergence of causal inference and language processing. Still, research on causality in NLP remains scattered across domains without unified definitions, benchmark datasets and clear articulations of the remaining challenges. In this survey, we consolidate research across academic areas and situate it in the broader NLP landscape. We introduce the statistical challenge of estimating causal effects, encompassing settings where text is used as an outcome, treatment, or as a means to address confounding. In addition, we explore potential uses of causal inference to improve the performance, robustness, fairness, and interpretability of NLP models. We thus provide a unified overview of causal inference for the computational linguistics community.
Few-shot Knowledge Graph (KG) completion is a focus of current research, where each task aims at querying unseen facts of a relation given its few-shot reference entity pairs. Recent attempts solve this problem by learning static representations of entities and references, ignoring their dynamic properties, i.e., entities may exhibit diverse roles within task relations, and references may make different contributions to queries. This work proposes an adaptive attentional network for few-shot KG completion by learning adaptive entity and reference representations. Specifically, entities are modeled by an adaptive neighbor encoder to discern their task-oriented roles, while references are modeled by an adaptive query-aware aggregator to differentiate their contributions. Through the attention mechanism, both entities and references can capture their fine-grained semantic meanings, and thus render more expressive representations. This will be more predictive for knowledge acquisition in the few-shot scenario. Evaluation in link prediction on two public datasets shows that our approach achieves new state-of-the-art results with different few-shot sizes.
The problem of Multiple Object Tracking (MOT) consists in following the trajectory of different objects in a sequence, usually a video. In recent years, with the rise of Deep Learning, the algorithms that provide a solution to this problem have benefited from the representational power of deep models. This paper provides a comprehensive survey on works that employ Deep Learning models to solve the task of MOT on single-camera videos. Four main steps in MOT algorithms are identified, and an in-depth review of how Deep Learning was employed in each one of these stages is presented. A complete experimental comparison of the presented works on the three MOTChallenge datasets is also provided, identifying a number of similarities among the top-performing methods and presenting some possible future research directions.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191