Shahab Bakhtiari
Shahab Bakhtiari researches visual perception and learning in biological brains and artificial neural networks. He uses deep learning to model learning and perception in the brain. After electrical engineering degrees in Tehran, he studied perceptual learning and visuomotor transformation at McGill, then held a postdoctoral fellowship at Mila. His work includes a self-supervised network that models dorsal and ventral visual pathways, a Patterns commentary on predictive coding and energy efficiency, and research on visual representation geometry. As of October 2026, he is an assistant professor at Université de Montréal, an associate academic member of Mila, and a principal investigator.
- Now at
- Mila and Université de Montréal
- Works
- 32
- Citations
- 152
- h-index
- 6
Publication numbers from OpenAlex, as of 14 October 2026.
Timeline
-
By 2012
Training in electrical engineering
He earned undergraduate and graduate degrees in electrical engineering at the University of Tehran before beginning neuroscience training at McGill.
-
2012–2018
Tracing perception and movement
From 2012 to 2018, he studied perceptual learning and visuomotor transformation in McGill’s Integrated Program in Neuroscience.
-
2019–2022
A postdoctoral bridge
At Mila, he held a postdoctoral fellowship from 2019 to 2022.
-
2021
Modeling two visual pathways
His NeurIPS paper trains a deep neural network with parallel pathways and a self-supervised predictive loss to model both dorsal and ventral visual processing in mouse cortex.
-
2022
Energy behind prediction
His Patterns commentary discusses evidence that predictive coding emerges in an artificial neural network optimized for energy efficiency.
-
2025
How representations change
He coauthors a study that compares representations in human visual cortex and artificial neural networks, and links their geometry to the computational objective.
-
As of October 2026
A faculty role in Montreal
He is an assistant professor in the Department of Psychology at Université de Montréal, an associate academic member of Mila, and a principal investigator. In his research, he studies visual perception and learning in biological brains and artificial neural networks. He uses deep learning to model learning and perception in the brain.
Sources and credits
Photo: snailab.ca; snail - Shahab Bakhtiari; creator not stated. License: Not stated by source. Source: snailab.ca. Cropped to a square and resized. Ask us to remove or replace this photo.
- Shahab BAKHTIARI - Département de psychologie - Université de Montréal
- Shahab
- Team
- The functional specialization of visual cortex emerges from training parallel pathways with self-supervised predictive learning
- Energy efficiency as a normative account for predictive coding
- [2509.13459] Why all roads don't lead to Rome: Representation geometry varies across the human visual cortical hierarchy
- Shahab Bakhtiari | Mila
- snail - Team