Intelligent Robot Learning Laboratory (IRL Lab) YunShu Du

yunshu

CONTACT INFORMATION:
YunShu Du
PhD Student, Computer Science
Email: yunshu.du@wsu.edu
Office: Dana Hall 3
Links: Personal Website


My Story

I am now a third year PhD student under the supervision of Dr. Matthew E. Taylor. From 2010 to 2012, I majored in software engineering at Wuhan University in China. I transferred to Eastern Michigan University to study computer science in my junior year. After two years of study, I obtained a bachelor degree of computer science in 2014.

Research

I’m interested in deep learning, agent mining, and applying machine learning algorithms to real-world problems

Current Projects

By: Yunshu Du and Matthew E. Taylor

WSU’s Recreation Center (the Rec) is among the most frequently visited campus facilities. However, students may prefer to avoid the Rec when it is most crowded. Our work aims to solve this problem by predicting how crowded the Rec will be at different times by leveraging the university’s CougCard system.

CougCard, the WSU official identification card, is used by all students when entering the Rec. This work used anonymized CougCard data from the Rec and applied data-driven techniques to analyze student exercise trends. A predictive decision tree model was successfully built to predict the peak hours at the Rec. A web-based application for the model is currently under construction with the goal of suggesting when the Rec will be more or less busy.

Our long term goal is to make students more (quantitatively) satisfied with their experience at the Rec and/or (quantitatively) increase the number of times they visit the Rec to exercise. Additionally, our system can assist Rec managers with shift scheduling and fitness event planning. Future work includes building personal fitness recommendations into the application and increasing the number of areas the application monitors and predicts crowdedness (e.g., the CUB’s food court). [1]

[1] [pdf] Yunshu Du and Matthew E. Taylor. Work In-progress: Mining the Student Data for Fitness . In Proceedings of the 12th International Workshop on Agents and Data Mining Interaction (ADMI) (at AAMAS), Singapore, May 2016.
[Bibtex]
@inproceedings{2016ADMI-Du,
author={Yunshu Du and Matthew E. Taylor},
title={{Work In-progress: Mining the Student Data for Fitness }},
booktitle={{Proceedings of the 12th International Workshop on Agents and Data Mining Interaction ({ADMI}) (at {AAMAS})}},
year={2016},
address={Singapore},
month={May},
abstract = {Data mining-driven agents are often used in applications such as waiting times estimation or traffic flow prediction. Such approaches often require large amounts of data from multiple sources, which may be difficult to obtain and lead to incomplete or noisy datasets. University ID card data, in contrast, is easy to access with very low noise. However, little attention has been paid to the availability of these datasets and few applications have been developed to improve student services on campus. This work uses data from CougCard, the Washington State University official ID card, used daily by most students. Our goal is to build an intelligent agent to improve student service quality by predicting the crowdedness at different campus facilities. This work in-progress focuses on the University Recreation Center, one of the most popular facilities on campus, to optimize students’ workout experiences.}
}

Videos & Other Media:

By: Yunshu DuGabriel V. de la Cruz Jr., James Irwin, and Matthew E. Taylor

As one of the first successful models that combines reinforcement learning technique with deep neural networks, the Deep Q-network (DQN) algorithm has gained attention as it bridges the gap between high-dimensional sensor inputs and autonomous agent learning. However, one main drawback of DQN is the long training time required to train a single task. This work aims to leverage transfer learning (TL) techniques to speed up learning in DQN. We applied this technique in two domains, Atari games and cart-pole, and show that TL can improve DQN’s performance on both tasks without altering the network structure. [1]

[1] [pdf] Yunshu Du, Gabriel V. de la Cruz Jr., James Irwin, and Matthew E. Taylor. Initial Progress in Transfer for Deep Reinforcement Learning Algorithms. In Proceedings of Deep Reinforcement Learning: Frontiers and Challenges workshop (at IJCAI), New York City, NY, USA, July 2016.
[Bibtex]
@inproceedings{2016DeepRL-Du,
author={Du, Yunshu and de la Cruz, Jr., Gabriel V. and Irwin, James and Taylor, Matthew E.},
title={{Initial Progress in Transfer for Deep Reinforcement Learning Algorithms}},
booktitle={{Proceedings of Deep Reinforcement Learning: Frontiers and Challenges workshop (at {IJCAI})}},
year={2016},
address={New York City, NY, USA},
month={July},
bib2html_pubtype={Refereed Workshop or Symposium},
abstract={As one of the first successful models that combines reinforcement learning technique with deep neural networks, the Deep Q-network (DQN) algorithm has gained attention as it bridges the gap between high-dimensional sensor inputs and autonomous agent learning. However, one main drawback of DQN is the long training time required to train a single task. This work aims to leverage transfer learning (TL) techniques to speed up learning in DQN. We applied this technique in two domains, Atari games and cart-pole, and show that TL can improve DQN’s performance on both tasks without altering the network structure.
}
}

News

Publications

2016

  • Yunshu Du, Gabriel V. de la Cruz Jr., James Irwin, and Matthew E. Taylor. Initial Progress in Transfer for Deep Reinforcement Learning Algorithms. In Proceedings of Deep Reinforcement Learning: Frontiers and Challenges workshop (at IJCAI), New York City, NY, USA, July 2016.
    [BibTeX] [Abstract] [Download PDF]

    As one of the first successful models that combines reinforcement learning technique with deep neural networks, the Deep Q-network (DQN) algorithm has gained attention as it bridges the gap between high-dimensional sensor inputs and autonomous agent learning. However, one main drawback of DQN is the long training time required to train a single task. This work aims to leverage transfer learning (TL) techniques to speed up learning in DQN. We applied this technique in two domains, Atari games and cart-pole, and show that TL can improve DQN’s performance on both tasks without altering the network structure.

    @inproceedings{2016DeepRL-Du,
    author={Du, Yunshu and de la Cruz, Jr., Gabriel V. and Irwin, James and Taylor, Matthew E.},
    title={{Initial Progress in Transfer for Deep Reinforcement Learning Algorithms}},
    booktitle={{Proceedings of Deep Reinforcement Learning: Frontiers and Challenges workshop (at {IJCAI})}},
    year={2016},
    address={New York City, NY, USA},
    month={July},
    bib2html_pubtype={Refereed Workshop or Symposium},
    abstract={As one of the first successful models that combines reinforcement learning technique with deep neural networks, the Deep Q-network (DQN) algorithm has gained attention as it bridges the gap between high-dimensional sensor inputs and autonomous agent learning. However, one main drawback of DQN is the long training time required to train a single task. This work aims to leverage transfer learning (TL) techniques to speed up learning in DQN. We applied this technique in two domains, Atari games and cart-pole, and show that TL can improve DQN’s performance on both tasks without altering the network structure.
    }
    }

  • Yunshu Du and Matthew E. Taylor. Work In-progress: Mining the Student Data for Fitness . In Proceedings of the 12th International Workshop on Agents and Data Mining Interaction (ADMI) (at AAMAS), Singapore, May 2016.
    [BibTeX] [Abstract] [Download PDF]

    Data mining-driven agents are often used in applications such as waiting times estimation or traffic flow prediction. Such approaches often require large amounts of data from multiple sources, which may be difficult to obtain and lead to incomplete or noisy datasets. University ID card data, in contrast, is easy to access with very low noise. However, little attention has been paid to the availability of these datasets and few applications have been developed to improve student services on campus. This work uses data from CougCard, the Washington State University official ID card, used daily by most students. Our goal is to build an intelligent agent to improve student service quality by predicting the crowdedness at different campus facilities. This work in-progress focuses on the University Recreation Center, one of the most popular facilities on campus, to optimize students’ workout experiences.

    @inproceedings{2016ADMI-Du,
    author={Yunshu Du and Matthew E. Taylor},
    title={{Work In-progress: Mining the Student Data for Fitness }},
    booktitle={{Proceedings of the 12th International Workshop on Agents and Data Mining Interaction ({ADMI}) (at {AAMAS})}},
    year={2016},
    address={Singapore},
    month={May},
    abstract = {Data mining-driven agents are often used in applications such as waiting times estimation or traffic flow prediction. Such approaches often require large amounts of data from multiple sources, which may be difficult to obtain and lead to incomplete or noisy datasets. University ID card data, in contrast, is easy to access with very low noise. However, little attention has been paid to the availability of these datasets and few applications have been developed to improve student services on campus. This work uses data from CougCard, the Washington State University official ID card, used daily by most students. Our goal is to build an intelligent agent to improve student service quality by predicting the crowdedness at different campus facilities. This work in-progress focuses on the University Recreation Center, one of the most popular facilities on campus, to optimize students’ workout experiences.}
    }