cadence
A brain that learns from live experience. System 1 by default: perception, memory, plasticity and action. Optional System 2 for self-observation and slow correc
What is cadence?
A brain that learns from live experience. System 1 by default: perception, memory, plasticity and action. Optional System 2 for self-observation and slow corrections.
Overview of cadence
What it is
Cadence is a brain that learns from live experience. It implements System 1 by default, which includes perception, memory, plasticity, and action. An optional System 2 adds self-observation and slow corrections. Cadence builds a simulated animal-like brain from simplified biological mechanisms. It is alpha research software.
Key features
- One continuing equilibrium brain that learns from live experience - Default System 1 with perception, plastic connections, memory, and action - Optional System 2 with observing cortical regions that feed back into the same neural graph - Uses reciprocal regions whose activity settles together, without a backward differentiation pass - Integrated learned world prediction, repair localized to what failed, and cheap stable operation (work in progress)
- Imagination function (
imagine) inspects responses without rewriting live experience - Supports explicit teaching, batched streams, and learning from every outcome via the lower-levelsteploop
Use cases
- Testing acquisition, retention, and recovery through a life - Target includes full-sentence language and reusable world understanding; a small fixed-label task is a control - Suitable for research in continuous learning, memory, and adaptive behavior
Getting started
Python 3.11+ and NumPy are required. Install the published release: sh python -m pip install cadence-net==0.85.1 Basic usage: python import numpy as np from cadence import Brain brain = Brain.compose(inputs=4, actions=2, arousal=True) observation = np.array([[1.0, 0.0, 0.0, 0.0]]) action = brain.live(observation) reward = (action == 0).astype(float) next_observation = np.array([[0.0, 1.0, 0.0, 0.0]]) action = brain.live(next_observation, reward=reward) To work from source, install the library repository root as an editable package: sh python -m pip install -e .
License and pricing
License: GPL-3.0. Open source.
Based on the project's public README and repository metadata.