Archive

Archive for December, 2013

Weekly QUEST Discussion Topics and News, 20 Dec

December 19, 2013 Leave a comment

Weekly QUEST Discussion Topics and News
20 Dec 2013
Our last meeting of the calendar year – we will take next week off –
Topics:
1.) Andres R will provide us a summary of the work he heard about while at the AFOSR cognition programs review.
2.) Igor T will provide us a review of some work he and Leonid P have done that might provide us a model to use to implement some of the situational based processing we’ve been discussing in our meetings
3.) Capt Amerika – has two articles he will make the group aware of
a. First one if ‘Dynamic change of Global and Local Information Processing in Propofol-Induced loss and recovery of consciousness’ – Monti et al from PLOS computational biology – “employ graph-theoretic measures and support vector machine classification to assess, in 12 healthy volunteers, the dynamic reconfiguration of functional connectivity during wakefulness, propofol-induced sedation and loss of consciousness, and the recovery of wakefulness. Our main findings, based on resting-state fMRI, are three-fold. First, we find that propofol-induced anesthesia does not bear differently on long-range versus short-range connections. Second, our multi-stage design dissociated an initial phase of thalamo-cortical and cortico-cortical hyperconnectivity, present during sedation, from a phase of cortico-cortical hypoconnectivity, apparent during loss of consciousness. Finally, we show that while clustering is increased during loss of consciousness, as recently suggested, it also remains significantly elevated during wakefulness recovery. Conversely, the characteristic path length of brain networks (i.e., the average functional distance between any two regions of the brain) appears significantly increased only during loss of consciousness, marking a decrease of global information-processing efficiency uniquely associated with unconsciousness. These findings suggest that propofol-induced loss of consciousness is mainly tied to cortico-cortical and not thalamo-cortical mechanisms, and that decreased efficiency of information flow is the main feature differentiating the conscious from the unconscious brain”
b. The second article was sent to me by Igor T – it addresses the failings of current approaches to AI – Hector Levesque thinks his computer is stupid—and that yours is, too. Siri and Google’s voice searches may be able to understand canned sentences like “What movies are showing near me at seven o’clock?,” but what about questions—“Can an alligator run the hundred-metre hurdles?”—that nobody has heard before? Any ordinary adult can figure that one out. (No. Alligators can’t hurdle.) But if you type the question into Google, you get information about Florida Gators track and field. Other search engines, like Wolfram Alpha, can’t answer the question, either. Watson, the computer system that won “Jeopardy!,” likely wouldn’t do much better. In a terrific paper just presented at the premier international conference on artificial intelligence, Levesque, a University of Toronto computer scientist who studies these questions, has taken just about everyone in the field of A.I. to task. He argues that his colleagues have forgotten about the “intelligence” part of artificial intelligence.

Weekly QUEST Discussion Topics and News 20 Dec

Advertisements

Better attachment for meeting topics, 13 Dec

December 12, 2013 Leave a comment
Categories: Uncategorized

Weekly QUEST Discussion Topics and News, Dec 13

December 12, 2013 Leave a comment

Below is the sequence of thoughts / ideas that will consume our meetings between now and the end of the year. What we seek is a robust solution to building robust human-computer symbiotic systems – and it is our premise that to achieve this we need to have computer solutions that mimic the dual system architecture – example see Stanovich and Evans – used by nature – thus we need a theory of consciousness – as cathy is out for a couple of weeks there will be a delay in getting the material posted to the VDL unless someone wants to step in to do that – we are in the CCR tomorrowas far as I know –
• THE ULTIMATE GOAL of a theory of consciousness is a simple and elegant set of fundamental laws, analogous to the fundamental laws of physics.
• From the Evans and Stanovich 2013 article – ‘dual process theories of higher cognition’: advancing the debate – key defining characteristic for Type 2 processing from Table 1 from that article
– Requires working memory
– Cognitive decoupling (allows for Hypothetical)
– Mental simulation
• It is the quest premise that the purpose of qualia representation is to have flexibility on the implication – flexible response – cognitive flexibility – achieved through decoupling and simulation – and the qualia are the parts of the Type 2 representation that are being attended to (part of working memory).
• QUEST definition of conscious is the generation of qualia – machine consciousness will be defined to be the generation of artificial qualia
– Artificial qualia will be defined as instantiations of computer implementations that are in compliance with the tenets in this discussion – done in working memory and cognitively decoupled and done in simulation
• Three QUEST tenets:
• 1.) Structural Coherence (could eventually roll this up into the two tenets below)
– Interaction – enough fidelity with reality (bits of awareness info) facilitate conscious driven interaction (situations are learned via interactions)
– There is a similarity measure applicable for the conscious representation – (color wheel example) – the similarity measure is between situations/qualia/chunks
• 2.) Situation based processing (situations as variables) – fundamental unit of conscious cognition
– Links (types of links, possibly capture similarity, meaning as what links are evoked – source of exformation – entities (situations) are defined based on how they are situated that is their meaning)
– Gists as key part of representation–low bandwidth representation –what is situated/simulated
– Multimodal – integrates multiple sensors representation into common framework – part of situated
– One quale at a time – for any aspect of the illusory cartesean theater (example invertible illusions)
– Qualia Theory of Relativity – only value (meaning) is in the relationships between (dictionary)
– Narrative based representation–situated in time/space/multiple modalities (plausible narratives compete)
– TD/BU – means to do context – rapid high level first –
– Types of Qualia types of situations – time as a quale, Affect as a quale, types of speech, ToM – (Evolving not static), aha and negative aha (means to know what is known and what is NOT known by the agent)
• Self – special type of qualia/situation (qualia self interacts with Continuity, unity, embodiment, sense of free will, reflection)
• 3.) Conscious representation of situations are done via simulation (cognitively decoupled)
– simulation is an organized body of knowledge that produces specific simulations of a situation’s instances
– Imagined past, imagined present, imagined future – cognitively decoupled
– Exformation (pattern completion inferring mechanism)
– Compression (infinite number of stimuli into a single quale, low bandwidth 50 bits/sec)
• One of the key ideas we’ve recently focused on is using the model of LaRue as an implementation of the Stanovich / Evans framework – to investigate using the model in a particular area some engineering decisions have to be made – for example in the secondary representation what will the java code in the agents be doing? How can we implement our tenets in that model.
• Taking as guidance our research into the importance of situational processing to Type 2 processing: One of the most potent factors in cognition is the background situation that frames a stimulus (also called context). In almost every task—from early perception to high-level reasoning—background situations have been shown to exert powerful effects on performance. Although a comprehensive review of situation effects has never been attempted, these effects are no doubt ubiquitous and substantial. *** it is our QUEST position that the fundamental unit of conscious cognition are situations ***
• Consider some prominent examples of the importance of situations / context.
– In language comprehension, texts can be incomprehensible when the relevant situation is not known (Bransford & Johnson, 1973; Sanford & Garrod, 1981).
– During conversations, situations are central to establishing common ground between speakers (Clark, 1992) and
– also in nonhuman communication (Smith, 1977).
• QUEST working definition of situation: A situation is any part the subjective internal representation of an agent which can be comprehended as a whole by that agent through defining how it interacts with or is related to other parts of the representation in that agent.
• Context – there is no reason to assume that the brain computes using numbers – although we have discussed the idea of the universal computing machine (Turing) it may be to accomplish what the brain does with perceptions when done with numbers would require the ‘infinite tape’
So I propose we investigate implementing situational based processing as a key aspect of the secondary representation computations in the LaRue model – we’ve discussed previously computing with Qualia:
Computational Theory of Perception: Computing with Qualia v2 Notes by Capt Amerika
Originally Nov 2007 modified Dec 2008 Adapted from work of Lofti Zadeh Int J Appl Math Comput Sci 2002, vol 12, no 3 307-324
• Goal – to replicate the ability developed in nature to accomplish a wide variety of task some without additional measurements (thus using exformation) or computations – manipulation of perceptions evoked by the measurements (this is how we define conscious reasoning)
• Manipulation of perceptions (reasoning) play a key role in the general purpose ability of nature to recognize patterns and make decisions
• Bring a better understanding of how nature does perception-based decision making.
• Where Prof Zadeh list examples of perceptions – “distance, size, weight, color, speed, time, direction, force, number, truth, likelihood and other characteristics of physical and mental objects” – ***it is our position that perceptions are NOT characteristics of the entities but are evoked by the entities and only exists in the internal representation of the critter perceiving them – see for example our definition of situation !***
• Certainly it would be our hope that we devise a method of generating subjective representations inside our digital computer – thus we would be using measurements to evoke those qualia inside our digital solutions *** and we will manipulate that internal representation – ‘consciously’ reason ***
• We hypothesize that the fundamental reason for the failure of our current ‘intelligent’ computational approaches is the “unavailability of a methodology for reasoning and computing with perceptions rather than measurements” – he was right spot on with this statement! And we believe perceptions are situations / qualia.
• In the Zadeh approach – he contends the failure is due to measurements being crisp whereas perceptions are fuzzy. We contend perceptions are qualia and even though fuzzy that formalism doesn’t or may NOT capture the essence of perceptions. *** it is the tying of the perceptions to attributes of the entities and total commitment to fuzzy only representation of what is being sensed is where I lose faith in the CWW – Zadeh approach ***
• What our representation of qualia will end up being – whether fuzzy or graph theoretic or category theoretic is TBD – but for now we assume they are subjective and temporal links between entities and GISTs!
• ***Those links are formed by taking sensory measurements as a stream and applying knowledge and memory to reason with the representation to generate the experienced hierarchical with multiple plausible narratives that compute the qualia at multiple levels of abstraction ***
• So the cornerstone of this discussion will be how to implement the ‘Link game’ as the fundamental computational process for the application using the LaRue model

WeeklyQUESTDiscussionTopicsandNews13Dec

NO QUEST MEETING 6 Dec

December 6, 2013 Leave a comment

Due to inclement weather conditions, WPAFB has been closed for the day, so we have been forced to cancel our QUEST meeting. We will plan on resuming our regular schedule next week.

Weekly QUEST Discussion Topics and News, 6 Dec 2013

December 5, 2013 Leave a comment

Weekly QUEST Discussion Topics and News
6 Dec 2013

Below is the sequence of thoughts / ideas that will consume our meetings between now and the end of the year. What we seek is a robust solution to building robust human-computer symbiotic systems – and it is our premise that to achieve this we need to have computer solutions that mimic the dual system architecture – example see Stanovich and Evans – used by nature – thus we need a theory of consciousness – as cathy is out for a couple of weeks there will be a delay in getting the material posted to the VDL unless someone wants to step in to do that – we are in the CCR tomorrowas far as I know –
• THE ULTIMATE GOAL of a theory of consciousness is a simple and elegant set of fundamental laws, analogous to the fundamental laws of physics.
• From the Evans and Stanovich 2013 article – ‘dual process theories of higher cognition’: advancing the debate – key defining characteristic for Type 2 processing from Table 1 from that article
– Requires working memory
– Cognitive decoupling (allows for Hypothetical)
– Mental simulation
• It is the quest premise that the purpose of qualia representation is to have flexibility on the implication – flexible response – cognitive flexibility – achieved through decoupling and simulation – and the qualia are the parts of the Type 2 representation that are being attended to (part of working memory).
• QUEST definition of conscious is the generation of qualia – machine consciousness will be defined to be the generation of artificial qualia
– Artificial qualia will be defined as instantiations of computer implementations that are in compliance with the tenets in this discussion – done in working memory and cognitively decoupled and done in simulation
• Three QUEST tenets:
• 1.) Structural Coherence (could eventually roll this up into the two tenets below)
– Interaction – enough fidelity with reality (bits of awareness info) facilitate conscious driven interaction (situations are learned via interactions)
– There is a similarity measure applicable for the conscious representation – (color wheel example) – the similarity measure is between situations/qualia/chunks
• 2.) Situation based processing (situations as variables) – fundamental unit of conscious cognition
– Links (types of links, possibly capture similarity, meaning as what links are evoked – source of exformation – entities (situations) are defined based on how they are situated that is their meaning)
– Gists as key part of representation–low bandwidth representation –what is
situated/simulated
– Multimodal – integrates multiple sensors representation into common framework – part of situated
– One quale at a time – for any aspect of the illusory cartesean theater (example invertible illusions)
– Qualia Theory of Relativity – only value (meaning) is in the relationships between (dictionary)
– Narrative based representation–situated in time/space/multiple modalities (plausible narratives compete)
– TD/BU – means to do context – rapid high level first –
– Types of Qualia types of situations – time as a quale, Affect as a quale, types of speech, ToM – (Evolving not static), aha and negative aha (means to know what is known and what is NOT known by the agent)
• Self – special type of qualia/situation (qualia self interacts with Continuity, unity, embodiment, sense of free will, reflection)
• 3.) Conscious representation of situations are done via simulation (cognitively decoupled)
– simulation is an organized body of knowledge that produces specific simulations of a situation’s instances
– Imagined past, imagined present, imagined future – cognitively decoupled
– Exformation (pattern completion inferring mechanism)
– Compression (infinite number of stimuli into a single quale, low bandwidth 50 bits/sec)
• One of the key ideas we’ve recently focused on is using the model of LaRue as an implementation of the Stanovich / Evans framework – to investigate using the model in a particular area some engineering decisions have to be made – for example in the secondary representation what will the java code in the agents be doing? How can we implement our tenets in that model.
• Taking as guidance our research into the importance of situational processing to Type 2 processing: One of the most potent factors in cognition is the background situation that frames a stimulus (also called context). In almost every task—from early perception to high-level reasoning—background situations have been shown to exert powerful effects on performance. Although a comprehensive review of situation effects has never been attempted, these effects are no doubt ubiquitous and substantial. *** it is our QUEST position that the fundamental unit of conscious cognition are situations ***
• Consider some prominent examples of the importance of situations / context.
– In language comprehension, texts can be incomprehensible when the relevant situation is not known (Bransford & Johnson, 1973; Sanford & Garrod, 1981).
– During conversations, situations are central to establishing common ground between speakers (Clark, 1992) and
– also in nonhuman communication (Smith, 1977).
• QUEST working definition of situation: A situation is any part the subjective internal representation of an agent which can be comprehended as a whole by that agent through defining how it interacts with or is related to other parts of the representation in that agent.
• Context – there is no reason to assume that the brain computes using numbers – although we have discussed the idea of the universal computing machine (Turing) it may be to accomplish what the brain does with perceptions when done with numbers would require the ‘infinite tape’
So I propose we investigate implementing situational based processing as a key aspect of the secondary representation computations in the LaRue model – we’ve discussed previously computing with Qualia:

Computational Theory of Perception: Computing with Qualia v2 Notes by Capt Amerika

Originally Nov 2007 modified Dec 2008 Adapted from work of Lofti Zadeh Int J Appl Math Comput Sci 2002, vol 12, no 3 307-324

• Goal – to replicate the ability developed in nature to accomplish a wide variety of task some without additional measurements (thus using exformation) or computations – manipulation of perceptions evoked by the measurements (this is how we define conscious reasoning)
• Manipulation of perceptions (reasoning) play a key role in the general purpose ability of nature to recognize patterns and make decisions
• Bring a better understanding of how nature does perception-based decision making.
• Where Prof Zadeh list examples of perceptions – “distance, size, weight, color, speed, time, direction, force, number, truth, likelihood and other characteristics of physical and mental objects” – ***it is our position that perceptions are NOT characteristics of the entities but are evoked by the entities and only exists in the internal representation of the critter perceiving them – see for example our definition of situation !***
• Certainly it would be our hope that we devise a method of generating subjective representations inside our digital computer – thus we would be using measurements to evoke those qualia inside our digital solutions *** and we will manipulate that internal representation – ‘consciously’ reason ***
• We hypothesize that the fundamental reason for the failure of our current ‘intelligent’ computational approaches is the “unavailability of a methodology for reasoning and computing with perceptions rather than measurements” – he was right spot on with this statement! And we believe perceptions are situations / qualia.
• In the Zadeh approach – he contends the failure is due to measurements being crisp whereas perceptions are fuzzy. We contend perceptions are qualia and even though fuzzy that formalism doesn’t or may NOT capture the essence of perceptions. *** it is the tying of the perceptions to attributes of the entities and total commitment to fuzzy only representation of
what is being sensed is where I lose faith in the CWW – Zadeh approach ***
• What our representation of qualia will end up being – whether fuzzy or graph theoretic or category theoretic is TBD – but for now we assume they are subjective and temporal links between entities and GISTs!
• ***Those links are formed by taking sensory measurements as a stream and applying knowledge and memory to reason with the representation to generate the experienced hierarchical with multiple plausible narratives that compute the qualia at multiple levels of abstraction ***
• So the cornerstone of this discussion will be how to implement the ‘Link game’ as the fundamental computational process for the application using the LaRue model

WeeklyQUESTDiscussionTopicsandNews6Dec