Aran Nayebi
Aran Nayebi is a machine learning researcher at Carnegie Mellon University who uses AI models to study how animals see, learn and act. He earned degrees in mathematics and computer science at Stanford and completed a neuroscience PhD there in 2022. At MIT, he led a 2023 study of models that predicted future events in videos and reproduced aspects of animal brain activity during mental simulation. He coauthored work on autonomous virtual zebrafish in 2025, and his 2026 AAAI paper examined mathematical limits on human–AI alignment. He leads the NeuroAgents Lab at Carnegie Mellon, where he is an assistant professor in the Machine Learning Department.
- Works
- 73
- Citations
- 1,710
- h-index
- 17
Publication numbers from OpenAlex, as of 27 October 2026.
Timeline
-
2015–2017
Stanford mathematics and computer science
He earns a BS in mathematics at Stanford University in 2015 and an MS in computer science there in 2017.
-
2018
Recurrent models of vision
In 2018, he coauthors Task-Driven Convolutional Recurrent Models of the Visual System. The team searches thousands of network designs to identify recurrent cells and feedback connections that help recognize objects.
-
2020
Identifying neural learning rules
In 2020, he coauthors Identifying Learning Rules From Neural Network Observables. The study trains classifiers on simulated neural network measurements to distinguish learning rules and guide experiments on how brains learn.
-
March 2022
Neuroscience PhD at Stanford
He completes a PhD in Stanford University’s Neurosciences Graduate Program in March 2022, co-advised by Dan Yamins and Surya Ganguli.
-
2023
Mental simulation at MIT
As a postdoctoral researcher in MIT’s K. Lisa Yang Integrative Computational Neuroscience Center, he leads a study presented at NeurIPS in 2023. The team trains models to predict future states from videos and finds that their activity resembles animal brain activity in Mental-Pong, a task that requires tracking a hidden ball.
-
2024
Career award and CMU appointment
In 2024, he receives a Burroughs Wellcome Fund Career Award at the Scientific Interface for “Building Integrative, Embodied Agents to Understand Natural Cognition.” He joins Carnegie Mellon University’s Machine Learning Department as an assistant professor in 2024.
-
2025
Autonomous agents from virtual zebrafish
In 2025, he coauthors Intrinsic Goals for Autonomous Agents: Model-Based Exploration in Virtual Zebrafish Predicts Ethological Behavior and Whole-Brain Dynamics. The team’s 3M-Progress method uses an internal drive to explore, producing virtual zebrafish that predict patterns of animal behavior and brain activity.
-
2026
Mathematical limits on AI alignment
His paper Intrinsic Barriers and Practical Pathways for Human-AI Alignment: An Agreement-Based Complexity Analysis appears at AAAI in 2026. It treats alignment as agreement among agents over multiple objectives and proves limits on the communication needed as the number of agents or objectives grows.
-
As of October 2026
NeuroAgents at Carnegie Mellon
As of October 2026, he leads Carnegie Mellon’s NeuroAgents Lab, which studies animal intelligence and develops AI systems that learn and adapt. The lab also studies how to keep autonomous AI under human control. He is an assistant professor in the Machine Learning Department, a core faculty member of the Neuroscience Institute, and holds a courtesy appointment in the Robotics Institute.
Career
- Since 2024Carnegie Mellon UniversityAssistant Professor, Lab Lead, NeuroAgents Lab
- 2022–2024Massachusetts Institute of TechnologyPostdoctoral Researcher
Education
- 2022Stanford UniversityPhD
- 2017Stanford UniversityMS
- 2015Stanford UniversityBS
Sources and credits
Photo: We have no published photo of Aran Nayebi yet. Send us a free-licensed photo.
- Aran Nayebi | MetaConscious Group
- [1807.00053] Task-Driven Convolutional Recurrent Models of the Visual System
- [2010.11765] Identifying Learning Rules From Neural Network Observables
- People - NeuroAILab
- About Me | Aran Nayebi
- The brain may learn about the world the same way some computational models do | MIT News | Massachusetts Institute of Technology
- Announcing the 2024 Career Awards at the Scientific Interface Recipients - Burroughs Wellcome Fund
- School of Computer Science < Carnegie Mellon University
- [2506.00138] Intrinsic Goals for Autonomous Agents: Model-Based Exploration in Virtual Zebrafish Predicts Ethological Behavior and Whole-Brain Dynamics
- Research | Aran Nayebi
- [2502.05934] Intrinsic Barriers and Practical Pathways for Human-AI Alignment: An Agreement-Based Complexity Analysis
- CMU Builds on Its Strengths To Advance NeuroAI | Mellon College of Science
