A5101457350- Reinforcement Learning in Robotics
- Auction Theory and Applications
- Advanced Clustering Algorithms Research
- Game Theory and Applications
- Multi-Criteria Decision Making
- Multi-Agent Systems and Negotiation
- Gene expression and cancer classification
- Advanced Decision-Making Techniques
- Cognitive Science and Mapping
- Bioinformatics and Genomic Networks
- Fuzzy Logic and Control Systems
- Optimization and Search Problems
- Advanced Algorithms and Applications
- Advanced Computational Techniques and Applications
- Vehicle License Plate Recognition
- MicroRNA in disease regulation
- Evaluation Methods in Various Fields
- Bach Studies and Logistics Development
- Rough Sets and Fuzzy Logic
- Text and Document Classification Technologies
- Evaluation and Optimization Models
- Face and Expression Recognition
- Advanced Numerical Analysis Techniques
- Graph Theory and Algorithms
- Collaboration in agile enterprises
University of Chinese Academy of Sciences
2023-2024
Nanjing University of Finance and Economics
2007-2024
Capital University of Economics and Business
2023
Qingdao University
2023
Kunming University of Science and Technology
2017-2021
Jilin Jianzhu University
2009
Abstract Background Clustering is a fundamental problem in statistics and has broad applications various areas. Traditional clustering methods treat features equally ignore the potential structure brought by characteristic difference of features. Especially cancer diagnosis treatment, several types biological are collected analyzed together. Treating these fails to identify heterogeneity both data itself, which leads incompleteness inefficacy current anti-cancer therapies. Objectives In this...
Dynamic pricing in electronic marketplaces is a basic problem commercial. In multiagent environments, the optimal policy of agent depends on policies other agents. This makes learning more problematic. paper proposes an efficient online algorithm, which integrates observed objective actions as well subjective inferential intention opponents. By establishing decision model agents and predicting their proposed price advance, becomes adaptive to its opponents can make good decisions long terms....
In electronic marketplaces automated and dynamic pricing is becoming increasingly popular. Agents that perform this task can improve themselves by learning from past observations, possibly using reinforcement techniques. several papers studied the use of Q-learning for modeling problem in marketplaces. But The extension (RL) to large state space has inevitably encountered curse dimensionality. Improving efficiency agent much more important practical application RL. To address pricing, we...
Graph partitioning (GP) is a classic problem that divides the node set of graph into densely-connected blocks. Following IEEE HPEC Challenge and recent advances in pre-training techniques (e.g., large-language models), we propose PR-GPT (Pre-trained & Refined ParTitioning) based on novel refinement paradigm. We first conduct offline deep learning (DGL) model small synthetic graphs with various topology properties. By using inductive inference DGL, one can directly generalize pre-trained...
Most of the machine learning techniques requires high level data integrity to achieve ideal performance. However, in real application scenario with complex conditions, especially social network, missing is such a general problem that has certain impact on effect behavior mining using learning. To dress mining, we proposed combine Multiple Imputation by Chained Equations (MICE) Random Forest algorithm, which applied imputation and then conducted model training predicting. Experiments were...
The problem of sellers pricing in electronic marketplaces can be ranked as a coordination games, the paper puts forward agentspsila belief revision model and learning algorithm which is based on similarity strategies games. By position-exchanging, each agent stands from viewpoint its opponent infers opponentspsila actions,. combines objective observed actions subjective inferred actions. Coordination assured by adjusting degree to 0 or 1. simulations results prove that our makes more...
Abstract Background: Clustering is a fundamental problem in statistics and has broad applications various areas. Traditional clustering methods treat features equally ignore the potential structure brought by characteristic difference of features. Especially cancer diagnosis treatment, several types biological are collected analyzed together. Treating these fails to identify heterogeneity both data itself, which leads incompleteness inefficacy current anti-cancer therapies. Results: In this...
Q-learning is a Reinforcement Learning (RL) model from the field of artificial intelligence, several papers studied use for modeling problem dynamic pricing in electronic marketplaces. But existing RL comes short achieving good result as amount exploration required often too costly. To address pricing, we take Bayesian model-based approach, framing transition function and reward MDP distributions, sampling technique action selection. The Beyesian approach accounts general...
Background: Decision-making trial and evaluation laboratory (DEMATEL) is a practical concise method to deal with the complicated socioeconomic system problems. However, there are two defects in original DEMATEL. On one hand traditional expert preference expressions can’t reflect hesitation flexibility of expert, on other determination group experts’ weight usually be expressed equivalent which scientificality behalf academic background, capability experience, risk so on. To solve above...