Fireside Chat
summary
Career arc and concrete contributions cover a decades-long mix of theory and applied work: a Science paper that resolves Berkeley grad admissions / Simpson's paradox with department-level conditioning; foundational results in semi-parametric theory that feed modern causal inference; bootstrap and robust-statistics developments tied to computation; a nonparametric p-value approach and an ENCODE reproducibility / producer method used across a consortium; skeptical, example-driven notes on LLM limits and ChatGPT hallucinations, and a push for theoreticians to shape complex, computational science.
AI-written summary · about
chapters (19)
- 0:00Welcome
- 1:48Early career and mentors
- 4:44Choosing statistics
- 8:31Influential figures
- 11:28Computation in research
- 12:57Proudest papers
- 18:10Robustness in statistics
- 23:32ENCODE contributions
- 26:04Theoreticians in AI
- 27:28Peter Bickel’s views on large language models
- 29:01Human thinking and AI
- 32:06Open questions in statistics
- 35:07Current interests
- 36:40ChatGPT experiences
- 40:20Publication culture
- 43:10Causal inference in economics
- 45:21Good theorems in ML
- 47:07Handling conflicting evidence
- 49:23Closing remarks
related talks
23mLive Fireside Chat
30mFireside Chat with Kelsey Hightower
27mCareer development panel: From being the only women in the room to empowering the next generation
40mAnti-disciplinary: doing academic research when you like too many things
01h 07mFireside Chat
37mFriday Night Live with Ethan and Aleksey
43mKeynote: Fireside Chat with Kelsey Hightower
01h 05mKeynote Panel: Building a career in GenAI