Prof Dev Talk 2024 – Dr. Gemma Roig – On aligning machine and visual cortex representations
On aligning machine and visual cortex representations
Presented by Dr. Gemma Roig
Abstract: I will present our current work on unveiling function specialization in the visual cortex using task-specific deep neural networks. I will also give an overview on our efforts in benchmarking brain-model alignment with the Argonauts Challenge, as well as developing an open framework coined NEt2Brain for facilitating the exchange between AI and brain science. Finally, I will discuss on the insights that we can gain with such endeavors and give some future perspectives on links between model inner interpretability and cognitive neuroscience.
On aligning machine and visual cortex representations
Presented by Dr. Gemma Roig
Abstract: I will present our current work on unveiling function specialization in the visual cortex using task-specific deep neural networks. I will also give an overview on our efforts in benchmarking brain-model alignment with the Argonauts Challenge, as well as developing an open framework coined NEt2Brain for facilitating the exchange between AI and brain science. Finally, I will discuss on the insights that we can gain with such endeavors and give some future perspectives on links between model inner interpretability and cognitive neuroscience.
![SIGIR 2024 T2.3 [pp] Axiomatic Causal Interventions for Reverse Engineering Relevance Computation](https://i.ytimg.com/vi/gbjXmMZMOe8/hqdefault.jpg "SIGIR 2024 T2.3 [pp] Axiomatic Causal Interventions for Reverse Engineering Relevance Computation")
