Scholar
Zhang-Wei Hong
Google Scholar ID: GZkyN4cAAAAJ
Massachusetts Institute of Technology
Citations & Impact
Citations
1,328
H-index
15
i10-index
19
Publications
20
Co-authors
15
list available
Publications
Resume (English only)
Research Experience
- Research Intern, MIT-IBM Watson AI Lab (Jun–Sep 2023), Advisor: Akash Srivastava.
- Research Intern, Microsoft Research Montreal (Jun–Oct 2022), Advisors: Romain Laroche and Remi Tachet des Combes.
- Research Intern, Preferred Networks (Jun–Oct 2019), Advisors: Prabhat Nagarajan and Dr. Guilherme Maeda.
- Research Intern, Appier (Feb–Jun 2019), Advisor: Prof. Min Sun.
- Visiting Researcher, Intelligent Autonomous Systems group, TU Darmstadt (Jul–Oct 2018), Advisor: Prof. Jan Peters.
- Research Assistant, ELSA Lab, National Tsing Hua University (Jul 2017–Mar 2020), Advisor: Prof. Chun-Yi Lee.
- Research Collaboration, Vision Science Lab, National Tsing Hua University (Oct 2016–Mar 2017), Advisor: Prof. Min Sun.
- Teaching Assistant, Taiwan NVIDIA Deep Learning Institute (2017–2018), Advisor: Prof. Chun-Yi Lee.
- Software Engineering Intern, Mediatek (Jul–Nov 2016), Advisor: Anthony Liu.
- Contract Software Engineer, Industrial Technology Research Institute (ITRI) (Oct–Dec 2015).
Background
- Research focuses on advancing reinforcement learning (RL) methods to overcome challenges in applying RL to computational discovery problems.
- Discovery problems span applications such as identifying materials that optimize power density and designing robot controllers for complex tasks.
- These problems require optimizing specific objectives using interaction data from systems with unknown dynamics in black-box settings.
- Believes RL is well-suited for discovery due to its trial-and-error learning paradigm.
- Key challenges include sparse reward signals limiting learning efficiency and lack of diversity in standard RL algorithms that aim for a single optimal solution.
