Metabolic cost of information processing in Poisson variational autoencoders
Hadi Vafaii, Jacob L. Yates Redwood Center for Theoretical Neuroscience, UC Berkeley · 2026
Read Metabolic cost of information processing in Poisson variational autoencoders by Hadi Vafaii, Jacob L. Yates Redwood Center for Theoretical Neuroscience, UC Berkeley (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2602-13421.txt.
Description
Brains compute under tight energy budgets, yet most theories of computation treat energy as free. This paper argues that variational free energy minimization under a Poisson assumption points toward an energy-aware theory. In the Poisson variational autoencoder, the KL term becomes proportional to prior firing rates. That creates an emergent metabolic cost that punishes high baseline activity.
That link couples coding rate, an abstract information quantity, to firing rate, a concrete biological variable. It trades fidelity against energy. Tests across KL weightings and latent sizes show raising the KL weight steadily increases sparsity and cuts spiking in the Poisson model, while a Gaussian control model stays unchanged. Hadi Vafaii leads the work with one co-author.
Catalog record
| CREATOR | Hadi Vafaii, Jacob L. Yates Redwood Center for Theoretical Neuroscience, UC Berkeley |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 96 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602-13421