Manifestro

DREAM

Dynamic Recall and Elastic Adaptive Memory — a neural architecture for continuous-time adaptation with surprise-driven plasticity.

Overview

DREAM is a continuous-time recurrent neural network that adapts during inference. Unlike static architectures, it modulates plasticity and integration speeds based on prediction error — enabling real-time learning without gradient updates.

82K
Parameters
99.9%
Reconstruction
0 steps
Adaptation
1.09×
Noise Robustness

Architecture

Surprise-Driven Plasticity

Learning rates modulated by prediction error surprise using Hebbian weight updates.

Liquid Time-Constants

Integration speeds adapt continuously based on input novelty and temporal context.

Fast Weights

Low-rank weight decomposition enables efficient meta-learning with minimal overhead.

Sleep Consolidation

Memory stabilization during low-surprise periods prevents catastrophic forgetting.

Benchmarks

ModelParametersImprovementAdaptation
DREAM82K99.9%0 steps
LSTM893K93.9%0 steps
Transformer551K92.6%0 steps

Applications

Online Learning

Real-time adaptation to streaming data

Speech Recognition

Acoustic modeling with temporal dynamics

Time Series

Forecasting with regime change detection

Few-Shot Learning

Rapid task adaptation

Anomaly Detection

Reconstruction-based outlier detection

Non-Stationary Data

Handle distribution shift

Get Started

Install DREAM and start building adaptive neural networks.