Diverse, top-k, and top-quality planning are concerned with the generation of sets of solutions to sequential decision problems. Previously this area has been the domain of classical planners that require a symbolic model of the problem instance. This paper proposes a novel alternative approach that uses Monte Carlo Tree Search (MCTS), enabling application to problems for which only a black-box simulation model is available. We present a procedure for extracting bounded sets of plans from pre-generated search trees in best-first order, and a metric for evaluating the relative quality of paths through a search tree. We demonstrate this approach on a path-planning problem with hidden information, and suggest adaptations to the MCTS algorithm to increase the diversity of generated plans. Our results show that our method can generate diverse and high-quality plan sets in domains where classical planners are not applicable.
相關內容
Markov Chain Monte Carlo methods for sampling from complex distributions and estimating normalization constants often simulate samples from a sequence of intermediate distributions along an annealing path, which bridges between a tractable initial distribution and a target density of interest. Prior work has constructed annealing paths using quasi-arithmetic means, and interpreted the resulting intermediate densities as minimizing an expected divergence to the endpoints. We provide a comprehensive analysis of this 'centroid' property using Bregman divergences under a monotonic embedding of the density function, thereby associating common divergences such as Amari's and Renyi's ${\alpha}$-divergences, ${(\alpha,\beta)}$-divergences, and the Jensen-Shannon divergence with intermediate densities along an annealing path. Our analysis highlights the interplay between parametric families, quasi-arithmetic means, and divergence functions using the rho-tau Bregman divergence framework of Zhang 2004,2013.
Learning Collaborative Information Dissemination with Graph-based Multi-Agent Reinforcement Learning
In modern communication systems, efficient and reliable information dissemination is crucial for supporting critical operations across domains like disaster response, autonomous vehicles, and sensor networks. This paper introduces a Multi-Agent Reinforcement Learning (MARL) approach as a significant step forward in achieving more decentralized, efficient, and collaborative solutions. We propose a Partially Observable Stochastic Game (POSG) formulation for information dissemination empowering each agent to decide on message forwarding independently, based on their one-hop neighborhood. This constitutes a significant paradigm shift from traditional heuristics based on Multi-Point Relay (MPR) selection. Our approach harnesses Graph Convolutional Reinforcement Learning, employing Graph Attention Networks (GAT) with dynamic attention to capture essential network features. We propose two approaches, L-DGN and HL-DGN, which differ in the information that is exchanged among agents. We evaluate the performance of our decentralized approaches, by comparing them with a widely-used MPR heuristic, and we show that our trained policies are able to efficiently cover the network while bypassing the MPR set selection process. Our approach is a first step toward supporting the resilience of real-world broadcast communication infrastructures via learned, collaborative information dissemination.
We consider the variable selection problem for two-sample tests, aiming to select the most informative variables to distinguish samples from two groups. To solve this problem, we propose a framework based on the kernel maximum mean discrepancy (MMD). Our approach seeks a group of variables with a pre-specified size that maximizes the variance-regularized MMD statistics. This formulation also corresponds to the minimization of asymptotic type-II error while controlling type-I error, as studied in the literature. We present mixed-integer programming formulations and develop exact and approximation algorithms with performance guarantees for different choices of kernel functions. Furthermore, we provide a statistical testing power analysis of our proposed framework. Experiment results on synthetic and real datasets demonstrate the superior performance of our approach.
In the domain of Federated Learning (FL) systems, recent cutting-edge methods heavily rely on ideal conditions convergence analysis. Specifically, these approaches assume that the training datasets on IoT devices possess similar attributes to the global data distribution. However, this approach fails to capture the full spectrum of data characteristics in real-time sensing FL systems. In order to overcome this limitation, we suggest a new approach system specifically designed for IoT networks with real-time sensing capabilities. Our approach takes into account the generalization gap due to the user's data sampling process. By effectively controlling this sampling process, we can mitigate the overfitting issue and improve overall accuracy. In particular, We first formulate an optimization problem that harnesses the sampling process to concurrently reduce overfitting while maximizing accuracy. In pursuit of this objective, our surrogate optimization problem is adept at handling energy efficiency while optimizing the accuracy with high generalization. To solve the optimization problem with high complexity, we introduce an online reinforcement learning algorithm, named Sample-driven Control for Federated Learning (SCFL) built on the Soft Actor-Critic (A2C) framework. This enables the agent to dynamically adapt and find the global optima even in changing environments. By leveraging the capabilities of SCFL, our system offers a promising solution for resource allocation in FL systems with real-time sensing capabilities.
Generating competitive strategies and performing continuous motion planning simultaneously in an adversarial setting is a challenging problem. In addition, understanding the intent of other agents is crucial to deploying autonomous systems in adversarial multi-agent environments. Existing approaches either discretize agent action by grouping similar control inputs, sacrificing performance in motion planning, or plan in uninterpretable latent spaces, producing hard-to-understand agent behaviors. Furthermore, the most popular policy optimization frameworks do not recognize the long-term effect of actions and become myopic. This paper proposes an agent action discretization method via abstraction that provides clear intentions of agent actions, an efficient offline pipeline of agent population synthesis, and a planning strategy using counterfactual regret minimization with function approximation. Finally, we experimentally validate our findings on scaled autonomous vehicles in a head-to-head racing setting. We demonstrate that using the proposed framework significantly improves learning, improves the win rate against different opponents, and the improvements can be transferred to unseen opponents in an unseen environment.
A successful commercial deployment of quantum key distribution (QKD) technologies requires integrating QKD links into existing fibers and sharing the same fiber networks with classical data traffic. To mitigate the spontaneous Raman scattering (SpRS) noise from classical data channels, several quantum/classical coexistence strategies have been developed. O-band solutions place the QKD channel in the O-band for lower SpRS noise but with the penalty of higher fiber loss and can rarely reach beyond 80 km of fiber; another method is C-band coexistence with attenuated classical channels, which sacrifices the performance of classical channels for lower SpRS noise. In this work, a time-interleaving technique is demonstrated to enable the co-propagation of quantum and classical channels in the C-band without sacrificing either performance. By embedding QKD pulses in the gaps between classical data frames, the quantum channel is isolated from SpRS noise in both wavelength and time domains. C-band co-propagation of a polarization-encoding decoy-state BB84 QKD channel with a 100 Gb/s QPSK channel is experimentally demonstrated with quantum bit error rate (QBER) of 1.12%, 2.04%, and 3.81% and secure key rates (SKR) of 39.5 kb/s, 6.35 kb/s, and 128 b/s over 20, 50, and 100 km fibers, respectively. These results were achieved with the presence of classical launch power up to 10 dBm, which is at least one order of magnitude higher than reported works. We also demonstrated the co-propagation of a QKD channel with eight classical channels with total launch power up to 18-dBm (9-dBm per channel), which is the highest power of classical channels reported in C-band coexistence works.
The accurate and interpretable prediction of future events in time-series data often requires the capturing of representative patterns (or referred to as states) underpinning the observed data. To this end, most existing studies focus on the representation and recognition of states, but ignore the changing transitional relations among them. In this paper, we present evolutionary state graph, a dynamic graph structure designed to systematically represent the evolving relations (edges) among states (nodes) along time. We conduct analysis on the dynamic graphs constructed from the time-series data and show that changes on the graph structures (e.g., edges connecting certain state nodes) can inform the occurrences of events (i.e., time-series fluctuation). Inspired by this, we propose a novel graph neural network model, Evolutionary State Graph Network (EvoNet), to encode the evolutionary state graph for accurate and interpretable time-series event prediction. Specifically, Evolutionary State Graph Network models both the node-level (state-to-state) and graph-level (segment-to-segment) propagation, and captures the node-graph (state-to-segment) interactions over time. Experimental results based on five real-world datasets show that our approach not only achieves clear improvements compared with 11 baselines, but also provides more insights towards explaining the results of event predictions.
Conventional methods for object detection typically require a substantial amount of training data and preparing such high-quality training data is very labor-intensive. In this paper, we propose a novel few-shot object detection network that aims at detecting objects of unseen categories with only a few annotated examples. Central to our method are our Attention-RPN, Multi-Relation Detector and Contrastive Training strategy, which exploit the similarity between the few shot support set and query set to detect novel objects while suppressing false detection in the background. To train our network, we contribute a new dataset that contains 1000 categories of various objects with high-quality annotations. To the best of our knowledge, this is one of the first datasets specifically designed for few-shot object detection. Once our few-shot network is trained, it can detect objects of unseen categories without further training or fine-tuning. Our method is general and has a wide range of potential applications. We produce a new state-of-the-art performance on different datasets in the few-shot setting. The dataset link is //github.com/fanq15/Few-Shot-Object-Detection-Dataset.
Object detection is considered as one of the most challenging problems in computer vision, since it requires correct prediction of both classes and locations of objects in images. In this study, we define a more difficult scenario, namely zero-shot object detection (ZSD) where no visual training data is available for some of the target object classes. We present a novel approach to tackle this ZSD problem, where a convex combination of embeddings are used in conjunction with a detection framework. For evaluation of ZSD methods, we propose a simple dataset constructed from Fashion-MNIST images and also a custom zero-shot split for the Pascal VOC detection challenge. The experimental results suggest that our method yields promising results for ZSD.
Multi-relation Question Answering is a challenging task, due to the requirement of elaborated analysis on questions and reasoning over multiple fact triples in knowledge base. In this paper, we present a novel model called Interpretable Reasoning Network that employs an interpretable, hop-by-hop reasoning process for question answering. The model dynamically decides which part of an input question should be analyzed at each hop; predicts a relation that corresponds to the current parsed results; utilizes the predicted relation to update the question representation and the state of the reasoning process; and then drives the next-hop reasoning. Experiments show that our model yields state-of-the-art results on two datasets. More interestingly, the model can offer traceable and observable intermediate predictions for reasoning analysis and failure diagnosis.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191