AIConfPaper
← Search

Michael Chang

13 accepted papers

2026

Smooth Human-Robot Shared Control for Autonomous Orchard Monitoring with UGVs (I)

Cheikh Melainine El Bou, Michele Focchi, Michael Chang, Marco Camurri, Karl Dietrich von Ellenrieder
ICRA 2026poster

Precision agriculture offers the opportunity to auto- mate routine or difficult tasks in orchards and vineyards, such as spraying or inspection, with Unmanned Ground Vehicles (UGV). In this context, human operators should be kept in the closed-loop control of the robot for safety and reliability. Th…

Cited by 0Scholar ↗
2025

INCLUDE: Evaluating Multilingual Language Understanding with Regional Knowledge

Angelika Romanou, Negar Foroutan, Anna Sotnikova, Sree Harsha Nelaturu, Shivalika Singh, Rishabh Maheshwary +51
ICLR 2025spotlight

The performance differential of large language models (LLM) between languages hinders their effective deployment in many regions, inhibiting the potential economic and societal value of generative AI tools in many communities. However, the development of functional LLMs in many languages (i.e., mult…

Cited by 9Source ↗PDF ↗Scholar ↗
2023

Hierarchical Abstraction for Combinatorial Generalization in Object Rearrangement

Michael Chang, Alyssa Li Dayan, Franziska Meier, Thomas L. Griffiths, Sergey Levine, Amy Zhang
ICLR 2023poster

Object rearrangement is a challenge for embodied agents because solving these tasks requires generalizing across a combinatorially large set of configurations of entities and their locations. Worse, the representations of these entities are unknown and must be inferred from sensory percepts. We pres…

Cited by 6Source ↗PDF ↗Scholar ↗
2023

Human-Timescale Adaptation in an Open-Ended Task Space

Jakob Bauer, Kate Baumli, Feryal Behbahani, Avishkar Bhoopchand, Nathalie Bradley-Schmieg, Michael Chang +21
ICML 2023oral

Foundation models have shown impressive adaptation and scalability in supervised and self-supervised learning problems, but so far these successes have not fully translated to reinforcement learning (RL). In this work, we demonstrate that training an RL agent at scale leads to a general in-context l…

Cited by 111Source ↗PDF ↗Scholar ↗
2023

Im-Promptu: In-Context Composition from Image Prompts

Bhishma Dedhia, Michael Chang, Jake Snell, Thomas L. Griffiths, Niraj Jha
NeurIPS 2023poster

Large language models are few-shot learners that can solve diverse tasks from a handful of demonstrations. This implicit understanding of tasks suggests that the attention mechanisms over word tokens may play a role in analogical reasoning. In this work, we investigate whether analogical reasoning c…

Cited by 3Source ↗PDF ↗Scholar ↗
2022

Object Representations as Fixed Points: Training Iterative Refinement Algorithms with Implicit Differentiation

Michael Chang, Thomas L. Griffiths, Sergey Levine
NeurIPS 2022accept

Current work in object-centric learning has been motivated by developing learning algorithms that infer independent and symmetric entities from the perceptual input. This often requires the use iterative refinement procedures that break symmetries among equally plausible explanations for the data, b…

Cited by 54Source ↗PDF ↗Scholar ↗
2021

Modularity in Reinforcement Learning via Algorithmic Independence in Credit Assignment

Michael Chang, Sid Kaushik, Sergey Levine, Tom Griffiths
ICML 2021oral

Many transfer problems require re-using previously optimal decisions for solving new tasks, which suggests the need for learning algorithms that can modify the mechanisms for choosing certain actions independently of those for choosing others. However, there is currently no formalism nor theory for…

Cited by 12Source ↗PDF ↗Scholar ↗
2020

Decentralized Reinforcement Learning: Global Decision-Making via Local Economic Transactions

Michael Chang, Sid Kaushik, S. Matthew Weinberg, Tom Griffiths, Sergey Levine
ICML 2020poster

This paper seeks to establish a framework for directing a society of simple, specialized, self-interested agents to solve what traditionally are posed as monolithic single-agent sequential decision problems. What makes it challenging to use a decentralized approach to collectively optimize a central…

Cited by 20Source ↗PDF ↗code ↗Scholar ↗
2019

Automatically Composing Representation Transformations as a Means for Generalization

Michael Chang, Abhishek Gupta, Sergey Levine, Thomas L. Griffiths
ICLR 2019poster

A generally intelligent learner should generalize to more complex tasks than it has previously encountered, but the two common paradigms in machine learning -- either training a separate learner per task or training a single learner for all tasks -- both have difficulty with such generalization beca…

Cited by 91Source ↗PDF ↗code ↗Scholar ↗
2019

Entity Abstraction in Visual Model-Based Reinforcement Learning

Rishi Veerapaneni, John D. Co-Reyes, Michael Chang, Michael Janner, Chelsea Finn, Jiajun Wu +2
CoRL 2019

We present OP3, a framework for model-based reinforcement learning that acquires object representations from raw visual observations without supervision and uses them to predict and plan. To ground these abstract representations of entities to actual objects in the world, we formulate an interactive

Cited by 0Source ↗PDF ↗code ↗Scholar ↗
2019

MCP: Learning Composable Hierarchical Control with Multiplicative Compositional Policies

Xue Bin Peng, Michael Chang, Grace Zhang, Pieter Abbeel, Sergey Levine
NeurIPS 2019poster

Humans are able to perform a myriad of sophisticated tasks by drawing upon skills acquired through prior experience. For autonomous agents to have this capability, they must be able to extract reusable skills from past experience that can be recombined in new ways for subsequent tasks. Furthermore,…

Cited by 247Source ↗PDF ↗code ↗Scholar ↗
2018

Relational Neural Expectation Maximization: Unsupervised Discovery of Objects and their Interactions

Sjoerd van Steenkiste, Michael Chang, Klaus Greff, Jürgen Schmidhuber
ICLR 2018poster

Common-sense physical reasoning is an essential ingredient for any intelligent agent operating in the real-world. For example, it can be used to simulate the environment, or to infer the state of parts of the world that are currently unobserved. In order to match real-world conditions this causal kn…

Cited by 326Source ↗PDF ↗code ↗Scholar ↗
2017

A Compositional Object-Based Approach to Learning Physical Dynamics

Michael Chang, Tomer Ullman, Antonio Torralba, Joshua Tenenbaum
ICLR 2017poster

We present the Neural Physics Engine (NPE), a framework for learning simulators of intuitive physics that naturally generalize across variable object count and different scene configurations. We propose a factorization of a physical scene into composable object-based representations and a neural net…

Cited by 530Source ↗Scholar ↗