Artificial neural networks
Recent articles
This paper changed my life: Appreciating John Hopfield’s brilliant neural network
In a 1982 paper, the Nobel laureate created his namesake recurrent neural network—work that taught Maria Geffen to always ground research questions in biology.
This paper changed my life: Appreciating John Hopfield’s brilliant neural network
In a 1982 paper, the Nobel laureate created his namesake recurrent neural network—work that taught Maria Geffen to always ground research questions in biology.
Kim Stachenfeld on the dance between neuroscience and artificial intelligence
As a researcher at both Google DeepMind and Columbia University, Stachenfeld offers cross-disciplinary insight into how to understand the brain.
Kim Stachenfeld on the dance between neuroscience and artificial intelligence
As a researcher at both Google DeepMind and Columbia University, Stachenfeld offers cross-disciplinary insight into how to understand the brain.
Explore more from The Transmitter
The illusion of AI consciousness: Lessons from human unconscious processing
Complex, goal-directed and even emotionally responsive behavior can unfold without awareness, providing a useful lens for interpreting artificial systems.
The illusion of AI consciousness: Lessons from human unconscious processing
Complex, goal-directed and even emotionally responsive behavior can unfold without awareness, providing a useful lens for interpreting artificial systems.
‘Push-pull’ recipe for neural wiring used in multiple brain regions
A versatile pair of proteins steers neurons toward their targets and helps establish the brain’s sensory maps, new studies suggest.
‘Push-pull’ recipe for neural wiring used in multiple brain regions
A versatile pair of proteins steers neurons toward their targets and helps establish the brain’s sensory maps, new studies suggest.
Reward-learning algorithm hardwired into dopamine circuit
The finding bolsters the canonical model of reward prediction error, which has come under scrutiny in recent years.
Reward-learning algorithm hardwired into dopamine circuit
The finding bolsters the canonical model of reward prediction error, which has come under scrutiny in recent years.