1. Ecorithms.
2. Prediction and adaptation.
3. The computable: not everything that can be defined can be computed. The Turing Paradigm
Robust computational models
The character of computational laws
Polynomial time computation
Possible ultimate limitations
Simple alogorithms with complicated behavior
The perceptron algorithm
4. Mechanistic explanations of nature : what might we look for?
5. The learnable : how can one draw general lessons from particular experiences?
Cognition
The problem of induction
Induction in an urn
Error control
Toward PAC learnability
PAC learnability
Occam: when to trust a hypothesis
Are there limits to learnability?
Teaching and learning
Learnable target pursuit
PAC learning as a basis of cognition
6. The evolvable : how can complex mechanisms evolve from simpler ones?
Is there a gap?
How can the gap be filled?
Does evolution have a target?
Evolvable target pursuit
Evolution versus learning
Evolution as a form of learning
Definition of evolvability
Extent and limits
Real-based evolution
Whyis this theory so different?
7. The deducible : how can one reason with imprecise concepts?
Reasoning
The need for reasoning even with the theoryless
The challenge of complexity
The challenge of brittleness
The challenge of semantics
The challenge of grounding
The mind's eye: a pinhole to the world
Robust logic: reasoning in an unknowable world
Thinking
8. Humans as ecorithms.
Introduction
Nature versus nurture
Naiveté
Prejudice and rush to judgement
Personalized truth
Personal feelings
Delusions of reason
Machine-aided humans
Is there something more?
9. Machines as ecorithms : why is artificial intelligence difficult to achieve?
Introduction
Machine learning
Artificial intelligence - where is the difficulty?
The artificial in artificial intelligence
Unsupervised learning
Artificial intelligence-where next?
Need we fear artificial intelligence?
Questions.
Science
A more strongly ecorithmic future
How to act?
Mysteries.