cv
You can download a PDF version. Although the cv is more updated on this website. I intend to disclose the whole journey of mine here, however it may look winding and strange to others. I want to build, one honest work a time, and then one item in this page. In the end it's life worth living.
Basics
| Name | Zhonghao He (何忠豪) |
| Label | AI Alignment and Human-AI Interaction Researcher |
| Affiliation | Leverhulme Center for Future Intelligence, University of Cambridge |
| zh378@cam.ac.uk | |
| Url | https://www.linkedin.com/in/hezhonghao |
| Summary | I build aligned and ethical LLMs for the benefits of many. I also build my skills to work on HCI, AI for science, interpretability. |
Education
-
2022.09 - PRESENT Cambridge, UK
Master
University of Cambridge
AI Ethics and Society
- Machine Learning Alignment Bootcamp
- AI Ethics
- AI Governance
- History of AI
- CS230 Deep Learning
- Mathematics for Computer Science
- ML Safety
- Discrete Mathematics
- CS234 Reinforcement Learning
- CS109 Probability for Computer Scientists
- CS106 Programming Methodology
- Mechanistic Interpretability
- Algorithms and Data Structure
-
2019.06 - 2019.09 Palo Alto, USA
Summer Student
Stanford University
Cognitive Science & Philosophy
- Mathematics Foundation of Computing
- Minds and Machines
- Introduction to Neuroscience
-
2014.08 - 2019.06 Shantou, China
Bachelor of Arts
Shantou University
English & Global Studies
- Machine Learning and relevant maths
- Research Methodology
- Linguistics
Projects
- 2024.10 - Present
LLM's influence on epistemic diversity and values
We are concerned with the problems of LLM-incured value lock-in, knowledge collapse, value collapse (as probable as model collapse since increasingly our discourses are mediated by AI systems and iterative training becomes more prevalence), with the consequence being more destructive. We created a team to build empirical demonstration, human subject experiments, simulation and interventions to address this set of problems we call 'AI influence'.
- Position: Project Co-founder
- Collaborators: Max Kleiman-Weiner, Tianyi (Alex) Qiu, Tejasveer Chugh
- 2023.12 - Present
Multilevel analytical framework for interpretability
Research on cognitive science and neuroscience to address interpretability challenges in ML.
- Position: Project Lead
- Goal: Publication in Transactional Machine Learning Research
- Senior authors: Adrian Weller, Grace W. Lindsay
- 2023.07 - 2023.10
Comprehensive Survey on AI Alignment
Survey paper on alignment research for newcomers.
- Focus: Interpretability challenges in ML
- Collaborators: Yaodong Yang, Jiaming Ji, Tianyi Qiu
- 2022.12 - 2023.03
Harms from agentic algorithmic systems
Research on safety and harms from agentic systems in AI.
- Highlight: Published paper cited by GPT-4 and high-profile AI safety reports
- 2021.06 - 2022.02
Stanford Existential Risks Initiative (SERI)
Research on China's AI global approach.
- Position: Research Fellow
Awards
- 2022.08
Open Philanthropy Graduate Scholarship
Open Philanthropy
A full scholarship for the master of study in AI Ethics and Society at the University of Cambridge.
- 2023.01
Manifund AI Research Scholarship
Manifund
A research scholarship to support work on neuroscience and mechanistic interpretability.
- 2017.02
Hong Kong Cyberport Creative Micro Funding
Hong Kong Cyberport Management Company Limited, wholly owned by the Hong Kong SAR Government
Early-stage startup funding to support work on Homeal, a sharing economy app about food and culture.
Skills
| Machematics | |
| Calculus | |
| Information Theory |
| ML/AI | |
| Convolution Neural Networks | |
| Transformers | |
| Autoencoders |
| Experimental | |
| Data Visualization | |
| Mechanitic Interpretability |
| Programming | |
| Python (advanced) | |
| R (intermediary) | |
| web stuff (intermediary) | |
| Matlab (basic) | |
| C/C++ (basic) |
Languages
| English | |
| Fluent |
| Chinese | |
| Native |
Interests
| Physical Activities | |
| Rowing | |
| Hiking |
| Other Interests | |
| Debate |