HOUSE_OVERSIGHT_027007.jpg
1.7 MB
Extraction Summary
1
People
2
Organizations
0
Locations
2
Events
0
Relationships
3
Quotes
Document Information
Type:
Scientific proposal / white paper page
File Size:
1.7 MB
Summary
This document is page 4 of a scientific proposal for the 'Structure of Mental Representation initiative.' It outlines a plan to create a scientific community centered on AI benchmarking challenges involving the comprehension of dynamic visual sequences and narratives. It also proposes the formation of a 'pilot team' to define goals and hypotheses for a cognitive architecture.
People (1)
| Name | Role | Context |
|---|---|---|
| Unknown Author | Proposer |
Uses 'I suggest' to outline the initiative; proposes the creation of a pilot team.
|
Organizations (2)
| Name | Type | Context |
|---|---|---|
| Structure of Mental Representation initiative |
The scientific project being proposed in the document.
|
|
| House Oversight Committee |
Identified via the footer stamp 'HOUSE_OVERSIGHT_027007'.
|
Key Quotes (3)
"The task of the Structure of Mental Representation initiative will at first consist in the creation of a large and competitive community, built around a set of benchmark problems."Source
HOUSE_OVERSIGHT_027007.jpg
Quote #1
"I suggest picking a set of problems that covers most of the above architectural requirements: the comprehension of dynamic visual sequences (movies) and narratives."Source
HOUSE_OVERSIGHT_027007.jpg
Quote #2
"To pull the Structure of Mental Representation initiative off the ground, we will need to create a pilot team, which is the object of this proposal."Source
HOUSE_OVERSIGHT_027007.jpg
Quote #3
Full Extracted Text
Complete text extracted from the document (2,406 characters)
4
Operations on these representations include reinforcement
learning and classification, perceptual top-down/bottom-up
processing, abstraction, analogy formation, associative and
syllogistic reasoning, planning, reflection, anticipation and
several modes of reorganization.
learning and classification, perceptual top-down/bottom-up
processing, abstraction, analogy formation, associative and
syllogistic reasoning, planning, reflection, anticipation and
several modes of reorganization.
Creating a scientific community
The task of the Structure of Mental Representation initiative will
at first consist in the creation of a large and competitive
community, built around a set of benchmark problems.
at first consist in the creation of a large and competitive
community, built around a set of benchmark problems.
I suggest picking a set of problems that covers most of the
above architectural requirements: the comprehension of
dynamic visual sequences (movies) and narratives. The
evaluation of comprehension may focus both on a discourse
level (asking the system general and specific questions about
the consumed visuals or narratives) and directly, by producing
dynamic renderings and depictions of knowledge represented
within the system. At least on the discourse level, a direct
comparison to the functional properties of child performance
at different cognitive stages is possible.
above architectural requirements: the comprehension of
dynamic visual sequences (movies) and narratives. The
evaluation of comprehension may focus both on a discourse
level (asking the system general and specific questions about
the consumed visuals or narratives) and directly, by producing
dynamic renderings and depictions of knowledge represented
within the system. At least on the discourse level, a direct
comparison to the functional properties of child performance
at different cognitive stages is possible.
The Comprehension
Challenge
Challenge
Turning the benchmark task into a regular competition allows a
direct comparison between models, and offers a strong
incentive for exchange of solutions among research groups. The
need to for broad solutions with given material and intellectual
resources will enforce a higher degree of the reuse of code and
ideas than we currently see in AI architectures (outside of
robotic soccer, where such competitions have turned out to be
highly successful).
direct comparison between models, and offers a strong
incentive for exchange of solutions among research groups. The
need to for broad solutions with given material and intellectual
resources will enforce a higher degree of the reuse of code and
ideas than we currently see in AI architectures (outside of
robotic soccer, where such competitions have turned out to be
highly successful).
By structuring the comprehension tasks into different levels
and sub-domains (such as basic language acquisition tasks,
grammar, perceptual and logical tasks, social reasoning/theory
of mind, affective evaluation, constructive problem solving, use
of analogies and metaphor etc.), we can formulate consistent
long-term goals and realistic short-term problems.
and sub-domains (such as basic language acquisition tasks,
grammar, perceptual and logical tasks, social reasoning/theory
of mind, affective evaluation, constructive problem solving, use
of analogies and metaphor etc.), we can formulate consistent
long-term goals and realistic short-term problems.
The Formation of a Pilot Team
To pull the Structure of Mental Representation initiative off the
ground, we will need to create a pilot team, which is the object
of this proposal. The pilot team has the following tasks:
ground, we will need to create a pilot team, which is the object
of this proposal. The pilot team has the following tasks:
- Bring a critically sized group of people together, with a
well-defined set of shared goals
well-defined set of shared goals
- Formulate an initial set of hypotheses
- Bring these in the context of an existing or hypothetical
cognitive architecture, so specifications can be derived
cognitive architecture, so specifications can be derived
HOUSE_OVERSIGHT_027007
Discussion 0
No comments yet
Be the first to share your thoughts on this epstein document