β-Vale networks: one state, many ways to compute
Explore two pools of neural drive through static, episodic, and spiking networks.
Explore the survey 14 min All fields By Field · Formal worlds Survey · 3 interactive demos Survey · 2 interactive demos Survey · 3 interactive demos With an interactive demo Keep crossing boundaries
AI & ML
What gets lost when a learning machine normalizes?
Explore machine learning, memory, and the information hidden by normalization.
3 readings · 3 subfieldsNeural networks
What information can a network keep inside its units?
Network architectures · Normalization · Spiking computation · Beta Valence Networks
Learning & decisions
What makes an additional distinction useful for prediction or action?
KL divergence · Reinforcement learning · Supervised learning · Unsupervised learning · Statistical theory
Time & adaptation
Can equal static predictions conceal different responses to change?
Time series analysis · Online learning · Memory
What a sleeping brain can rehearse
Explore how new observations, fading evidence, and rehearsal change a memory.
Explore the survey 11 minβ-Vale networks: one state, many ways to compute
Explore two pools of neural drive through static, episodic, and spiking networks.
Explore the survey 14 minStrong clocks, a quiet chorus
Discover how strong cellular clocks can cancel the collective rhythm.
Read the article 12 min