@flight-gnc.bsky.social
Reposted
To understand RNN computations, we globally linearized their hidden-unit activity using high-dimensional linear-Gaussian state-space models (SSMs).
High-d SSMs (latents dim > obs dim) have great performance, and are interpretable through tools from dynamical systems and control theory.
High-d SSMs (latents dim > obs dim) have great performance, and are interpretable through tools from dynamical systems and control theory.
July 27, 2025 at 9:31 PM
To understand RNN computations, we globally linearized their hidden-unit activity using high-dimensional linear-Gaussian state-space models (SSMs).
High-d SSMs (latents dim > obs dim) have great performance, and are interpretable through tools from dynamical systems and control theory.
High-d SSMs (latents dim > obs dim) have great performance, and are interpretable through tools from dynamical systems and control theory.
At the end of the talk addressing scaling they seem to indicate a role for their architecture as a Gibbs sampling pipeline/seed for a GAN running (deterministically) on GPU
extropic.ai
extropic.ai
Extropic | Home
Building thermodynamic computing hardware that is radically more energy efficient than GPUs.
extropic.ai
November 4, 2025 at 6:46 PM
At the end of the talk addressing scaling they seem to indicate a role for their architecture as a Gibbs sampling pipeline/seed for a GAN running (deterministically) on GPU
extropic.ai
extropic.ai
Trick or treat?
October 20, 2025 at 3:20 PM
Trick or treat?
Reposted
@andykeller.bsky.social @kempnerinstitute.bsky.social presented “Flow Equivariant Cybernetics”, a blueprint for agents that learn through continuous feedback with their environment.
October 15, 2025 at 1:55 PM
@andykeller.bsky.social @kempnerinstitute.bsky.social presented “Flow Equivariant Cybernetics”, a blueprint for agents that learn through continuous feedback with their environment.
Reposted
This year’s meeting will bring together leading researchers in #neuroscience, #robotics, #machinelearning, #controltheory, and #theoreticalneuroscience to explore how behaviour and intelligence emerge from complex feedback loops linking brains, bodies, and environments.
October 15, 2025 at 10:44 AM
This year’s meeting will bring together leading researchers in #neuroscience, #robotics, #machinelearning, #controltheory, and #theoreticalneuroscience to explore how behaviour and intelligence emerge from complex feedback loops linking brains, bodies, and environments.
