CV

Publications

METRICS 2.0: A machine-learning based optimization system for IC design

Hashemi, S., Ho, C. T., Kahng, A. B., Liu, H. Y., & Reda, S. (2018). METRICS 2.0: A machine-learning based optimization system for IC design. In Workshop on Open-Source EDA Technology (p. 21).

SVM Learning for GFIS Trimer Health Monitoring in Helium-Neon Ion Beam Microscopy

Kahng, A. B., Liu, H. Y., Park, C., Pichumani, R., & Saul, L. SVM Learning for GFIS Trimer Health Monitoring in Helium-Neon Ion Beam Microscopy., Advanced Process Control Conference, 2019

Offline Reinforcement Learning with Munchausen Regularization

Liu, Hsin-Yu, Bharathan Balaji, Rajesh Gupta, and Dezhi Hong. " Offline Reinforcement Learning with Munchausen Regularization " Offline Reinforcement Learning Workshop at Neural Information Processing Systems, 2021 (NeurIPS 2021).

Safe HVAC Control via Batch Reinforcement Learning

Liu, H. Y., Balaji, B., Gao, S., Gupta, R., & Hong, D. (2022, May). Safe HVAC Control via Batch Reinforcement Learning. In 2022 ACM/IEEE 13th International Conference on Cyber-Physical Systems (ICCPS) (pp. 181-192). IEEE.

B2RL: An open-source Dataset for Building Batch Reinforcement Learning

Liu, Hsin-Yu, Xiaohan Fu, Bharathan Balaji, Rajesh Gupta, and Dezhi Hong. "B2RL: an open-source dataset for building batch reinforcement learning." In Proceedings of the 9th ACM International Conference on Systems for Energy-Efficient Buildings, Cities, and Transportation, pp. 462-465. 2022.

Rule-based Policy Regularization for Reinforcement Learning-based Building Control

Education

• Ph.D in Computer Engineering, University of California San Diego, 2023 (expected)
• M.S. in Optoelectronics, National Taiwan University, 2008
• B.S. in Physics, National Central University, 2005

Work experience

• Sep. 2018 ~ Now: Graduate Student Researcher
  • University of California - San Diego (La Jolla, CA, USA)
    • Deep Reinforcement Learning research: ** Online/Offline RL policy regularization ** Open-source dataset for offline-RL ** Incorporating existing policy with RL ** Offline-to-Online RL
    • Building HVAC control with RL
• Summer 2020: Machine Learning Intern-Software Engineer
  • Cadence Design Systems Inc. (Milpitas, CA, USA)
    • Create functions for automatically generating partitions to accelerate design space exploration:
    • Include most of the critical paths distribution in entire design
    • Partitions with reduced design area and instance numbers by a factor around 1/3 to 1/4
    • Prototyping with both Machine Learning clustering (DBSCAN) and heat map grid-based algorithms
• 2013 ~ 2018: Principal Application Engineer
  • Cadence Design Systems Inc.
    • Develop customized GUIs/scripts using Cadence SKILL
    • Mask data preparation for foundry N14/N28 reticle frames
    • Host customer training and workshops
    • Co-work with R&D for functionality improvement, bug-fix and performance enhancement
• 2009 ~ 2013: OPC R&D Engineer
  • Taiwan Semiconductor Manufacturing Company (TSMC)
    • N20 interconnection layers OPC model build-up, OPC and DPT recipe optimization. (ProGen, Proteus)
    • Path-finding and optimization of double patterning issues: Odd-loop, VCMS, pattern loading, etc.
    • Aerial/resist image simulations for design rule & process robustness evaluation. (Brion Tachyon, S-Litho, ICWB)
    • Resolution enhancement researches: Forbidden pitch, combined source of annular and quasar, SMO, etc.( S-Litho)
    • Flow automation, such as pattern density calculator, DPT-to-OPC flow connector, etc. (Perl)
    • VCMS (Via-Correlated Metal Separation): BEOL dual damascene LELE cross-layer DPT overlay reduction scheme
    • Rule deck creation and optimization: LOP/MRC/XOR/CRC, etc. (Calibre DRC)
    • Test mask creations: PMK/OMK/VMK/DMK, etc. (Tcl, Laker)
    • Rule-based OPC for interconnection layers restricted rule

Skills

• Progamming Language:
  • Python / Tcl/ Perl/ SKILL (Cadence)
• Software: ** Machine Learning/Data Science: pytorch, tensorflow, numpy, matplotlib ** VLSI/OPC: Cadence Virtuoso/PVS, Synopsys Proteus/ProGen/IC Workbench/DFM, Mentor SVRF, Brion Tachyon, etc.

Talks

SVM Learning for GFIS Trimer Health Monitoring in Helium-Neon Ion Beam Microscopy

October 22, 2019

Oral Presentation at Advanced Process Control Conference (APC) 2019, San Antonio, TX, USA

Offline Reinforcement Learning with Munchausen Regularization

December 14, 2021

Poster Session at Offline Reinforcement Learning Workshop at Neural Information Processing Systems, 2021 (NeurIPS 2021), Virtual Only (Remote Event)

Safe HVAC Control via Batch Reinforcement Learning

May 04, 2022

Oral Presentation at International Conference on Cyber-Physical Systems (ICCPS 2022), Milan, Italy (Remote Event)

B2RL: An open-source Dataset for Building Batch Reinforcement Learning

November 08, 2022

Oral presentation at Third ACM SIGEnergy Workshop on Reinforcement Learning for Energy Management in Buildings & Cities (RLEM) (in conjunction with ACM BuildSys’22), Boston, Massachusetts, USA

Rule-based Policy Regularization for Reinforcement Learning-based Building Control

June 20, 2023

Oral presentation at The 14th ACM International Conference on Future Energy Systems (ACM e-Energy 2023), Orlando, Florida, USA

Teaching

Cadence Design Systems Inc.

UCSD ECE 260A VLSI Digital System Algorithm and Architecture

Service and leadership

• Taipei animal shelter volunteer