Researcher. Musician. Outdoor Enthusiast.
Brandon Trabucco

Brandon is currently a Ph.D. student at CMU in the Machine Learning Department advised by Professor Ruslan Salakhutdinov, and was an undergraduate at UC Berkeley working with Professor Sergey Levine. Brandon's research includes reasoning, generative models, and agents. He produces electronic music and considers how machine learning research can enhance creative workflows.

Brandon is an NDSEG Fellow supported in part by the DoD. His research is published in top ML conferences, and featured as oral presentations at the ICLR Foundation Models Workshop and CVPR Embodied AI Workshop, where his work MaSS won the Two-Phase Rearrangement Challenge.

Research

Effective Data Augmentation With Diffusion
Effective Data Augmentation With Diffusion Models Brandon Trabucco, Kyle Doherty, Max Gurinas, Ruslan Salakhutdinov ICLR 2023 Foundation Models Workshop (oral) NeurIPS 2023 Foundation Models Workshop (oral) NeurIPS 2023 Synthetic Data Workshop (oral) ICLR 2024 [pdf] [code] [talk]
3D Mapping and Semantic Search
A Simple Approach for Visual Room Rearrangement: 3D Mapping and Semantic Search Brandon Trabucco, Gunnar Sigurdsson, Robinson Piramuthu, Gaurav S. Sukhatme, Ruslan Salakhutdinov CVPR 2022 Embodied AI Workshop (oral) ICLR 2023 [pdf] [code] [talk]
AnyMorph
AnyMorph: Learning Transferable Policies By Inferring Agent Morphology Brandon Trabucco, Mariano Phielipp, Glen Berseth ICML 2022 [pdf] [talk] [blog]
Offline Model-Based Optimization
Design-Bench: Benchmarks for Data-Driven Offline Model-Based Optimization Brandon Trabucco*, Xinyang Geng*, Aviral Kumar, Sergey Levine ICML 2022 [pdf] [talk]
Offline Model-Based Optimization
Conservative Objective Models For Effective Model-Based Optimization Brandon Trabucco*, Aviral Kumar*, Xinyang Geng, Sergey Levine ICML 2021 [pdf] [talk]
Variational Order Inference
Discovering Autoregressive Orderings with Variational Inference Xuanlin Li*, Brandon Trabucco*, Dong Huk Park, Michael Luo, Sheng Shen, Trevor Darrell, Yang Gao ICLR 2021 [pdf] [talk]
Neural Program Synthesis
Synthetic Datasets for Neural Program Synthesis Richard Shin, Neel Kant, Kavi Gupta, Christopher Bender, Brandon Trabucco, Rishabh Singh, Dawn Song ICLR 2019 [pdf]

Music

Brandon Trabucco ยท New World