Intelligence as resource efficiency
Mathematical Neuroscience conference
2025-12-12
2025-12-12
Abstract
I start by asking what could it mean to understand the brain. My answer is that we need to understand the nontrivial thing that only the brain can do ("intelligence", although that's a risky word). I argue for a definition of intelligence that includes resource efficiency as a core part, and suggest that this may be why machine learning approaches with such vast resources produce intelligence so unlike human intelligence. I illustrate with examples from multimodal perception and modularity, and discuss how this might be useful for our ongoing work in heterogeneity, delays and neuromodulation.
Slides
Links
Related people
Related publications
2026
-
Ghosh M, Goodman DFM (2026)
Partial recurrence enables robust and efficient computation.
Communications AI & Computing
2025
-
Sun P, Achterberg J, Su Z, Goodman DFM, Akarca D
Exploiting heterogeneous delays for efficient computation in low-bit neural networks.
Preprint -
Anil S, Goodman DFM, Ghosh M (2025)
Fusing multisensory signals across channels and time.
PLoS Computational Biology -
Béna G, Goodman DFM (2025)
Dynamics of specialization in neural modules under resource constraints.
Nature Communications
2024
-
Ghosh M, Béna G, Bormuth V, Goodman DFM (2024)
Nonlinear fusion is optimal for a wide class of multisensory tasks.
PLoS Computational Biology
2021
-
Perez-Nieves N, Leung VCH, Dragotti PL, Goodman DFM (2021)
Neural heterogeneity promotes robust learning.
Nature Communications
Categories