Memory and learning
Keeping information, revisiting it, retrieving it and reusing earlier experience.
14 problemsLearning and intelligence
Human cognition and machine learning face different questions about learning, generalization, and goals.
Subjects: Brain and cognition · Machine learning
Explore this collection
Reasoning and decisions
Judgment, value, self-monitoring and the control of thought.
8 problemsMovement and the body
Learning movements, sensing a body, experiencing control and responding to touch.
8 problemsReading and numbers
Persistent reading and arithmetic difficulties raise different questions about learning and development.
2 problemsProblems in this collection
7 problemsHuman general intelligence
Performance across different cognitive tasks is often related. Which brain mechanisms help explain that shared variation and the many abilities that do not reduce to it?
Continual learning
An AI system can learn new tasks and lose skills it already had. How can it keep useful knowledge while adapting with limited memory and computation?
Out-of-distribution generalization
A model can predict well on familiar data and fail when conditions change. Which changes can it handle reliably, and how can we know?
Mechanistic interpretability
We can inspect a neural network's numbers without understanding its computation. Can explanations become reliable enough to predict and check its behavior?
AI alignment
Alignment asks how AI systems can reliably follow intended goals and constraints, including in situations different from those used in training.
Working memory maintenance
We can hold information briefly after it disappears. Which neural processes keep it available for the next action?
Biological credit assignment
A behavioral outcome depends on many neurons. How does learning change the particular cells and connections responsible for it?