2024 BAIR Graduate Directory – The Berkeley Artificial Intelligence Research Blog

Advisor(s): Ion Stoica

Research Blurb: My analysis curiosity lies broadly within the subject of Machine Learning and Artificial Intelligence. During my PhD I’ve centered on Environment Generation/ Curriculum Learning strategies for coaching Autonomous Agents with Reinforcement Learning. Specifically, I work on strategies that algorithmically generates numerous coaching environments (i.e., studying situations) for autonomous brokers to enhance generalization and pattern effectivity. Currently, I’m engaged on Large Language Model (LLM) based mostly autonomous brokers.
Jobs Interested In: Research Scientist, ML Engineer

Advisor(s): Laurent El Ghaoui

Research Blurb: My analysis delves into the theoretical points of deep implicit fashions, starting with a unified “state-space” illustration that simplifies notation. Additionally, my work explores numerous coaching challenges related to deep studying, together with issues amenable to convex and non-convex optimization. In addition to theoretical exploration, my analysis extends the potential functions to varied downside domains, together with pure language processing, and pure science.
Jobs Interested In: Research Scientist, Applied Scientist, Machine Learning Engineer

Advisor(s): Masayoshi Tomizuka, Wei Zhan

Research Blurb: My analysis focuses on machine studying and management algorithms for the difficult process of autonomous racing in Gran Turismo Sport. I leverage my background in Mechanical Engineering to find how machine studying and model-based optimum management can create protected, high-performance management techniques for robotics and autonomous techniques. A specific emphasis of mine has been methods to leverage offline datasets (e.g. human participant’s racing trajectories) to tell higher, extra pattern environment friendly management algorithms.
Jobs Interested In: Research Scientist and Robotics/Controls Engineer

Advisor(s): David Wagner

Research Blurb: I’m broadly desirous about the safety and security points of machine studying techniques. Most of my earlier works are within the area of adversarial machine studying, notably adversarial examples and robustness of machine studying algorithms. More lately, I’m enthusiastic about rising safety and privateness dangers on giant language fashions.
Jobs Interested In: Research scientist

Advisor(s): Alison Gopnik

Research Blurb: Eliza Kosoy works on the intersection of kid improvement and AI with Prof. Alison Gopnik. Her work contains creating evaluative benchmarks for LLMs rooted in little one improvement and learning how kids and adults use GenAI fashions reminiscent of ChatGPT/Dalle and kind psychological fashions about them. She’s an intern at Google engaged on the AI/UX staff and beforehand with the Empathy Lab. She has printed in Neurips, ICML, ICLR, Cogsci and cognition. Her thesis work created a unified digital atmosphere for testing kids and AI fashions in a single place for the needs of coaching RL fashions. She additionally has expertise constructing startups and STEM {hardware} coding toys.
Jobs Interested In: Research Scientist (little one improvement and AI), AI security (specializing in kids), User Experience (UX) Researcher (specializing in combined strategies, youth, AI, LLMs), Education and AI (STEM toys)

Advisor(s): Alexandre Bayen

Research Blurb: Under the mentorship of Prof. Alexandre Bayen, Fangyu focuses on the applying of optimization strategies to multi-agent robotic techniques, notably within the planning and management of automated automobiles.
Jobs Interested In: Faculty, or analysis scientist in management, optimization, and robotics

Advisor(s): Jacob Steinhardt, Moritz Hardt

Research Blurb: My analysis focus is in machine studying for protein modeling. I work on bettering protein property classification and protein design, in addition to understanding what totally different protein fashions study. I’ve beforehand labored on sequence fashions for DNA and RNA, and benchmarks for evaluating the interpretability and equity of ML fashions throughout domains.
Jobs Interested In: Research scientist

Advisor(s): Alexandre Bayen

Research Blurb: My thesis work has specialised in reinforcement studying for autonomous automobiles, specializing in enhancing decision-making and effectivity in utilized settings. In future work, I’m keen to use these rules to broader challenges throughout domains like pure language processing. With my background, my goal is to see the direct influence of my efforts by contributing to modern AI analysis and options.
Jobs Interested In: ML analysis scientist/engineer

Advisor(s): Bin Yu, Song Mei

Research Blurb: I’m desirous about growing a greater foundational understanding of deep studying and bettering sensible techniques, utilizing each theoretical and empirical methodology. Currently, I’m particularly desirous about bettering the effectivity of enormous fashions by learning methods to correctly scale hyperparameters with mannequin dimension.
Jobs Interested In: Research Scientist

Advisor(s): Pieter Abbeel and Trevor Darrell

Research Blurb: My work entails RL, BC, studying from people, and utilizing common sense basis mannequin reasoning for agent studying. I’m enthusiastic about language agent studying, supervision, alignment & robustness.
Jobs Interested In: Research scientist

Advisor(s): Laura Waller

Research Blurb: My analysis is on computational imaging, notably the space-time modeling for dynamic scene restoration and movement estimation. I additionally work on optical microscopy strategies, optimization-based optical design, occasion digital camera processing, novel view rendering.
Jobs Interested In: Research scientist, postdoc, school

Advisor(s): Ken Goldberg

Research Blurb: Imitation studying and reinforcement studying algorithms that scale to giant robotic fleets performing manipulation and different complicated duties.
Jobs Interested In: Research Scientist

Advisor(s): Stuart Russell

Research Blurb: My analysis focuses on making language fashions safe, strong and protected. I even have expertise in imaginative and prescient, planning, imitation studying, reinforcement studying, and reward studying.
Jobs Interested In: Research scientist

Advisor(s): Joseph Gonzalez

Research Blurb: Gorilla LLM – Teaching LLMs to make use of instruments (https://gorilla.cs.berkeley.edu/); LLM Execution Engine: Guaranteeing reversibility, robustness, and minimizing blast-radius for LLM-Agents included into consumer and enterprise workflows; POET: Memory sure, and vitality environment friendly fine-tuning of LLMs on edge gadgets reminiscent of smartphones and laptops (https://poet.cs.berkeley.edu/).
Jobs Interested In: Research Scientist

Advisor(s): Trevor Darrell, Joseph Gonzalez

Research Blurb: I work on bettering the reliability and security of multimodal fashions. My focus has been on localizing and lowering hallucinations for imaginative and prescient + language fashions, together with measuring and utilizing uncertainty and mitigating bias. My pursuits lay in making use of options to those challenges in precise manufacturing situations, quite than solely in tutorial environments.
Jobs Interested In: Applied analysis scientist in generative AI, security, and/or accessibility

Advisor(s): Pieter Abbeel

Research Blurb: My analysis lies in robotics, machine studying, and pc imaginative and prescient, with the first objective of studying generalizable robotic expertise from two angles: (1) Learning structured world fashions with spatial and temporal abstractions. (2) Pre-training visible illustration and expertise to allow information switch from Internet-scale imaginative and prescient datasets and simulators.
Jobs Interested In: Faculty, or analysis scientist

Advisor(s): Michael I. Jordan, Yi Ma

Research Blurb: My analysis pursuits are broadly in idea and observe of reliable machine studying, together with interpretability, privateness, and robustness.
Jobs Interested In: Faculty