Seeing Is Believing: Brain-Inspired Modular Training for Mechanistic Interpretability
We introduce Brain-Inspired Modular Training (BIMT), a method for making neural networks more modular and interpretable. Inspired by brains, BIMT embeds neurons in a geometric space and augments the loss function with a cost proportional to the length of each neuron connection. This is inspired by t...
Main Authors: | Liu, Ziming, Gan, Eric, Tegmark, Max |
---|---|
Format: | Article |
Published: |
Multidisciplinary Digital Publishing Institute
2024
|
Online Access: | https://hdl.handle.net/1721.1/153423 |
Similar Items
-
Seeing Is Believing: Brain-Inspired Modular Training for Mechanistic Interpretability
by: Ziming Liu, et al.
Published: (2023-12-01) -
Seeing is believing.
by: Buckingham, S
Published: (2003) -
Seeing is believing
by: Enrique Amaya
Published: (2016-10-01) -
Seeing is believing
by: Sergey Krasikov, et al.
Published: (2022-02-01) -
Deepfakes, Seeing is Believing?
by: Claudia López Frías
Published: (2023-12-01)