Optimizing the Spatio-Temporal Resource Search Problem with Reinforcement Learning (GIS Cup)

We consider a setting with an evolving set of requests for transportation from an origin to a destination before a deadline and a set of agents capable of servicing the requests. In this setting, an authority assigns agents to requests such that the average idle time of the agents is minimized. An example is the scheduling of taxis (agents) to meet incoming passenger requests for trips while ensuring that the taxis are empty as little as possible. We address the problem of spatial-temporal demand forecasting and competitive supply (SOUP) in two steps. First, we build a granular model that provides spatial-temporal predictions of requests. Specifically, we propose a Spatial-Temporal Graph Convolutional Sequential Learning (ST-GCSL) model that predicts requests across locations and time slots. Second, we provide means of routing agents to request origins while avoiding competition among the agents. In particular, we develop a demand-aware route planning (DROP) algorithm that considers both the spatial-temporal predictions and the supply-demand state. We report on extensive experiments with real-world data that offer insight into the performance of the solution and show that it is capable of outperforming the state-of-the-art proposals.

show abstract

“…• TripBandAgent [8]: TripBandAgent optimizes the taxi search strategy using reinforcement learning (RL).…”

Section: Methods For Comparisonmentioning

confidence: 99%

SOUP: Spatial-Temporal Demand Forecasting and Competitive Supply

Zheng

Ming

et al. 2021

IEEE Trans. Knowl. Data Eng.

View full text Add to dashboard Cite

show abstract

“…• SmartAgent [25]: SmartAgent uses non-negative matrix factorization (NMF) to model and predict the spatiotemporal distributions of requests, and then chooses destinations using a greedy heuristic. • TripBandAgent [9]: TripBandAgent optimizes the taxicabs search strategy by using reinforcement learning (RL).…”

Section: A Experimental Settingsmentioning

confidence: 99%

SOUP: Spatial-Temporal Demand Forecasting and Competitive Supply

Zheng

Ming

et al. 2020

Preprint

View full text Add to dashboard Cite

We consider a setting with an evolving set of requests for transportation from an origin to a destination before a deadline and a set of agents capable of servicing the requests. In this setting, an assignment authority is to assign agents to requests such that the average idle time of the agents is minimized. An example is the scheduling of taxis (agents) to meet incoming requests for trips while ensuring that the taxis are empty as little as possible. In this paper, we study the problem of spatial-temporal demand forecasting and competitive supply (SOUP). We address the problem in two steps. First, we build a granular model that provides spatial-temporal predictions of requests. Specifically, we propose a Spatial-Temporal Graph Convolutional Sequential Learning (ST-GCSL) algorithm that predicts the service requests across locations and time slots. Second, we provide means of routing agents to request origins while avoiding competition among the agents. In particular, we develop a demand-aware route planning (DROP) algorithm that considers both the spatial-temporal predictions and the supplydemand state. We report on extensive experiments with realworld and synthetic data that offer insight into the performance of the solution and show that it is capable of outperforming the state-of-the-art proposals.

show abstract

“…Ever since the human society has entered the era of big data, the quantity and type of digital learning resources on the Internet are increasing exponentially [1][2][3][4][5]. In front of such massive amount of learning resources, inevitably, students might get lost in the sea of knowledge or drift from the main topic from time to time, so the requirement of students for learning resource retrieval is growing and changing [6][7][8][9][10][11][12][13][14]. When a student has just get started with a certain course or come into contact with the knowledge of a certain field, he or she does not have the ability to analyze the relevance between the professional knowledge and the learning resources, so it's impossible for the student to accurately describe the learning resources in need [15][16][17].…”

Section: Introductionmentioning

confidence: 99%

A New Learning Resource Retrieval Method Based on Multi-knowledge Association Mining

Xiao¹,

Yang²,

Zhang³

2023

Int. J. Emerg. Technol. Learn.

View full text Add to dashboard Cite

Ever since the human society has entered the era of big data, the quantity and type of digital learning resources on the Internet are increasing exponentially, and the requirement of students for learning resource retrieval is on the rise. However, the existing methods of learning resource retrieval generally overlook the overall knowledge systems that the students have already possessed, so it’s impossible for them to predict the students’ learning path or perform deviation adjustment. In view of these issues, this paper aims to study a new learning resource retrieval method based on multi-knowledge association mining. At first, the paper introduces the application of the Knowledge Graph Embedding (KGE) technology in learning resource retrieval, proposes the problem of learning resource retrieval, and points out the goal of learning resource retrieval. Then, a breadth-first soft-matching search algorithm is introduced to attain the multiple association paths between students and learning resources, and a retrieval module is constructed based on association path to further learn the association paths between students and learning resources within the framework of feature learning, and to predict the probability of interactions between students and learning resources. At last, this paper evaluates the association paths and uses experimental results to verify the validity of the proposed learning resource retrieval method.

show abstract

Optimizing the Spatio-Temporal Resource Search Problem with Reinforcement Learning (GIS Cup)

Cited by 13 publications

References 3 publications

SOUP: Spatial-Temporal Demand Forecasting and Competitive Supply

SOUP: Spatial-Temporal Demand Forecasting and Competitive Supply

SOUP: Spatial-Temporal Demand Forecasting and Competitive Supply

A New Learning Resource Retrieval Method Based on Multi-knowledge Association Mining

Contact Info

Product

Resources

About