The hypotheses behind Soar
Next: How does Soar work? Problem spaces.
Up: No Title
Previous: Is Soar the unified theory of cognition?
Back: to main list of student notes
See Artificial Intelligence volume 33 page 58 (1987). Some of
these points will become clearer from the examples.
- Physical symbol system hypothesis. A general intelligence
must be realised with a symbolic system. I described what this meant in
the first lecture.
- Goal structure hypothesis. Control in a general intelligence
is maintained by manipulating symbolic representations of goals. For
a simple example, the goal-stack in Young's seriation PS, last week.
- Uniform elementary-representation hypothesis. There is a
single elementary representation for declarative knowledge. The term
declarative is usually opposed with procedural. Declarative
knowledge can be inspected and decoded; procedural knowledge is
``implicit'', skill-type knowledge, and can only be obeyed. In one
slogan, ``declarative knowledge is what you read; procedural knowledge
is what you run'', using these words in the computing sense.
- Problem space hypothesis. Problem spaces are the fundamental
organisational unit of all goal-directed behaviour.
- Production system hypothesis. Production systems are the
appropriate method of encoding all long-term knowledge.
- Universal-subgoaling hypothesis. Any decision can be an
object of goal-oriented attention.
- Automatic-subgoaling hypothesis. All goals are created
dynamically in response to impasses and are generated automatically by
the architecture. Unlike the classical production systems, where goals
are created by rules, and the architecture doesn't have any special
mechanisms for them.
- Control-knowledge hypothesis. Any decision can be controlled
by indefinite amounts of knowledge, both domain- dependent and
-independent.
- Weak-method hypothesis. The weak methods form the basic
methods of intelligence.
- Weak-method emergence hypothesis. The weak methods arise
directly from the system responding based on its knowledge of the task.
- Uniform-learning hypothesis. Goal-based chunking is the
general learning mechanism.
Next: How does Soar work? Problem spaces.
Up: No Title
Previous: Is Soar the unified theory of cognition?
Back: to main list of student notes
Jocelyn Ireson-Paine
Wed Feb 14 23:45:33 GMT 1996