• Machine learning researcher.
  • Goal: to understand and develop safe and beneficial autonomous systems.
  • Currently: Allen Institute for AI
    Priors: 🤗, Ph.D. Berkeley AI; Cornell `17; Intern at DeepMind, Facebook AI.

About me

Hello! I am a machine learning researcher at AI2.

Previously, I was at HuggingFace bootstrapping an RLHF team. Before that, I finished my PhD at the University of California, Berkeley, Department of Electrical Engineering and Computer Sciences, advised by Professor Kristofer Pister in the Berkeley Autonomous Microsystems Lab, and pseudo-advised by Roberto Calandra at Meta AI Research!

Hello! I am a machine learning researcher at AI2.

Previously, I was at HuggingFace bootstrapping an RLHF team. I recently finished my PhD at the University of California, Berkeley studying the intersection of robotics and machine learning. I was a member of the Department of Electrical Engineering and Computer Sciences, advised by Professor Kristofer Pister in the Berkeley Autonomous Microsystems Lab, and pseudo-advised by Roberto Calandra at Meta AI Research! I am actively involved in outreach and inclusion efforts and an advocate for mental health -- I was the EEGSA wellness chair and founder of the UC Berkeley Equal Access to Application Assistance program.

Prior to UC Berkeley, I was a proud member of Cornell Electrical and Computer Engineering 2017 where I learned to do research with the Lab of Plasma Studies and the SonicMEMs Lab. I bring my research foundation in hardware, models, and physics to the data-driven world of machine learning. At Cornell, I was a part of Cornell Lightweight Rowing.

I am happy to be a product of The Ocean State.

Nathan Lambert is a Research Scientist at the Allen Institute for AI focusing on RLHF. Previously, he helped build an RLHF research team at HuggingFace. He received his PhD from the University of California, Berkeley working at the intersection of machine learning and robotics. He was advised by Professor Kristofer Pister in the Berkeley Autonomous Microsystems Lab and Roberto Calandra at Meta AI Research. He was lucky to intern at Facebook AI and DeepMind during his Ph.D. Nathan was was awarded the UC Berkeley EECS Demetri Angelakos Memorial Achievement Award for Altruism for his efforts to better community norms.

I like to try and have fun between my many projects. You can find me on Strava, I also happen to be a brand ambassador for Picky Bars. I actively track my health, cook, and read (recipe and book pages in construction).

Recent Papers

Camels in a Changing Climate: Enhancing LM Adaptation with Tulu 2
H. Ivison and Y. Wang et al.
[arxiv]
November 20, 2023
The Alignment Ceiling: Objective Mismatch in Reinforcement Learning from Human Feedback
Nathan Lambert, Roberto Calandra
[arxiv]
October 31, 2023
Zephyr: Direct Distillation of LM Alignment
HuggingFace H4 Team
[arxiv]
October 25, 2023

Recent Talks

15min History of Reinforcement Learning and Human Feedback
[recording]
December 15, 2023
Objective Mismatch in Reinforcement Learning from Human Feedback
AI2
[recording]
August 29, 2023

Other

News

  • Sept 2023: I launched my podcast with Thomas Krendl Gilbert -- The Retort.
  • July 2023: I was on a couple podcasts about Llama 2 and LLM evaluation, see more here.
  • Summer 2023: I'll be heading to FAccT in Chicago and ICML in Hawaii to present tutorials on RLHF (the former more ethics focused, slides here, the latter on data and technical details, slides here).
  • January 2023: We shared a new paper on the fundamentals of machine learning systems: Measuring Data.
  • November 15 2022: I've joined the board of technical advisors at The Farama Foundation to improve open-source RL infrastructure
  • October 1 2022: we open-sourced a fun new tool for building embodied AI environments at HuggingFace: Simulate.
  • June 11 2022: I'm co-organizing a workshop on Building Accountable and Transparent RL at RLDM.
  • June 1 2022: I started my job at HuggingFace 🤗.
  • May 2022: I defended my thesis and finished my Ph.D.
  • April 2022: We released a new paper on documentation for RL systems -- Reward Reports. [link]
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
Writing
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.

Robotics

Intelligent & novel devices to interact with the physical world.

A conceptual rendering of novel microrobot flight trajectories.
A conceptual rendering of novel microrobot flight trajectories.

Machine Learning

The science of using data to decide in the presence of uncertainty.

The optimization landscape with Bayesian Optimization.

Society

Making sure the stakeholders of automation are in the conversation.

A diagram depicting existing fields of socio-technical inquiry in AI
A diagram depicting existing fields of socio-technical inquiry in AI