Archive for August, 2014

Weekly QuEST Discussion Topics and News, 29 Aug

August 28, 2014 Leave a comment

QuEST 29 Aug 2014:

1.)  The first topic has to do with the current debate on ISIS / ISIL.  It reminded me of a set of discussions that have happened over the last couple of years in the QuEST meetings and resulted in us putting together a ‘think piece’ on Fighting an Adaptable foe.  Specifically the common issues in fighting in cyber, fighting the war on cancer and the fight against terrorism.  In the area of the fight against cancer the basic idea — still in the experimental stages — is that cancer cells cannot turn into a lethal tumor without the cooperation of other cells nearby. That may be why autopsies repeatedly find that most people who die of causes other than cancer have at least some tiny tumors in their bodies that had gone unnoticed.  *** in fact confirms matt’s brothers observation – and the lung cancer observation – that found as many lung cancers in nonsmokers although clearly more smokers die from lung cancer ***According to current thinking, the tumors were kept in check, causing no harm. … It also may mean that cancers grow in part because normal cells surrounding them allowed them to escape. It also means that there might be a new way to think about treatment: cancer might be kept under control by preventing healthy cells around it from crumbling*** this is the provide security and safety strategy approach to asymmetric war ***…“Think of it as this kid in a bad neighborhood,” said Dr. Susan Love, a breast cancer surgeon and president of the Dr. Susan Love Research Foundation. “You can take the kid out of the neighborhood and put him in a different environment and he will behave totally differently.” … if there is interest we can revisit this ‘think piece’ to see if our newer QuEST ideas can impact differently now.

2.)   The second topic has to do with the discussion last week on how we do NOT think the solution to a general purpose artificial intelligence that can respond acceptably to the unexpected query is to learn the representation it is to learn/adapt the parameters of a simulation.  We are reminded in our interactions with Prof Geman – We have these quotes from his presentations he gave us:

The mind’s eye

  • The brain simulates
  • Representations must be nearly literal
  • We don’t learn representations; we learn the parameters of simulation (“strong priors”)



  • Nonparametric learning may have little or nothing to do with biological learning (ontogenetic & phylogenetic)
  • The advantages of simulation would explain the striking growth of the neocortex
  • The homogeneity of the cortex suggests repeatable and scalable rules of composition
  • Image understanding might be more a matter of constructing a scene model than of computing a classification


And with the work we discussed last week at QuEST that is all about classification / localization being where the big boys (google / facebook) are focused I think we are on an interesting path with QuEST … what I would like to discuss is finding relevant publications that attempt to attack the issue of the difference in learning the parameters for a simulation versus learning a representation?  How does this solve the Biederman problem?  And the answer to the Jared question – what are the parameters of the simulation? (to me they are the qualia – the vocabulary of conscious thought)


3.)  That brings us to the third topic – another Prof we’ve interacted with that inspired us to continue down the path of the conscious representation is a simulation versus a projection of sensory data – Prof Barsalou – a key attribute of simulation is the pattern completion inferencing – I would like to present his work that provides an interesting explanation of mirror neurons related to simulation – Mirroring as Pattern Completion Inferences within Situated Conceptualizations – … The classic account of mirroring is that it results from mirror neurons, namely, neurons that have both motor and perceptual tunings. Mirror neurons not only become active when an action is performed, but also when it is perceived.  Because these neurons become active during the perception of an action, they ground the perception in action simulation. An alternative account constitutes the thesis developed here: Mirroring is a special case of a basic cognitive process common across species, namely, Pattern Completion Inferences …  within Situated Conceptualizations (PCIwSC). According to PCIwSC, the brain is a situation processing architecture (Barsalou, 2003, 2009; Barsalou et al., 2003; Wilson-Mendenhall et al., 2011; Yeh and Barsalou, 2006).

Weekly QuEST Discussion Topics and News 29 Aug

Categories: Uncategorized

Weekly QuEST Discussion Topics and News, 22 Aug

August 21, 2014 Leave a comment

QuEST 22 Aug 2014

There are several news stories that we need to cover – the first is the recent LSCRC – large scale visual recognition challenge:

Started in 2010 by Stanford, Princeton and Columbia University scientists, the Large Scale Visual Recognition Challenge this year drew 38 entrants from 13 countries. The groups use advanced software, in most cases modeled loosely on the biological vision systems, to detect, locate and classify a huge set of images taken from Internet sources like Twitter. The contest was sponsored this year by Google, Stanford, Facebook and the University of North Carolina.

Contestants run their recognition programs on high-performance computers based in many cases on specialized processors called G.P.U.s, for graphic processing units.

This year there were six categories based on object detection, locating objects and classifying them. Winners included the National University of Singapore, the Oxford University, Adobe Systems, the Center for Intelligent Perception and Computing at the Chinese Academy of Sciences, as well as Google in two separate categories.

Accuracy almost doubled in the 2014 competition and error rates were cut in half, according to the conference organizers.

… This year performance took a big leap …

Despite the fact that contest is based on pattern recognition software that can be “trained” to recognize objects in digital images, the contest itself is made possible by the Imagenet database, an immense collection of more than 14 million images that have been identified by humans. The Imagenet database is publicly available to researchers at

In the five years that the contest has been held, the organizers have twice, once in 2012 and again this year, seen striking improvements in accuracy, accompanied by more sophisticated algorithms and larger and faster computers.

… This year almost all of the entrants used a variant of an approach known as a convolutional neural network, an approach first refined in 1998 by Yann LeCun, a French computer scientist who recently became director of artificial intelligence research at Facebook.

“This is LeCun’s hour,” said Gary Bradski, an artificial intelligence researcher who was the founder of Open CV, a widely used machine vision library of software tools. Convolutional neural networks have only recently begun to have impact because of the sharply falling cost of computing, he said, “In the past there were a lot of things people didn’t do because no one realized there would be so much inexpensive computing power available.”

The accuracy results this year improved to 43.9 percent, from 22.5 percent, and the error rate fell to 6.6 percent, from 11.7 percent, according to Olga Russakovsky, a Stanford University graduate researcher who is the lead organizer for the contest. Since the Imagenet Challenge began in 2010, the classification error rate has decreased fourfold, she said.

… “Human-level understanding is much deeper than machine image classification,” she said. “I can easily find a image that will fool the algorithm and I can’t do it with humans, but we’re making significant progress.”

Although machines have made great progress in object recognition, they are only taking baby steps in what scientists describe as “scene understanding,” the ability to comprehend what is happening in an image in human language.

“I really believe in the phrase that ‘a picture is worth a thousand words,’ not a thousand disconnected words,” said Dr. Li. ”It’s the ability to tell a complete story. That is the holy grail *** meaning making **

This last piece is where we want to discuss – where we have been many times before – what is ‘meaning making’ and how it is agent centric – and how does QuEST play in this space

Next there was a couple of articles on Big Data (one focused on healthcare and one on ‘data wrangling’) the places where QuEST and Big Data merge might be in these areas – in both cases we need to understand the role of the human/computer agents.

The last news article I want to hit briefly is the ‘man playing the violin while undergoing brain surgery’ – we have hit related topics recently when discussing whether consciousness can initiate action or not (also we’ve discussed the Penfield work).

Also I want to briefly hit a recent article that our colleague Sandy V brought to our attention on narratives and expertise. Modeling the Function of Narrative in Expertise by W. Korey MacDougall, Robert L. West, and Christopher Genovesi

• The use of narrative is ubiquitous in the development, exercise, and communication of expertise.

• Expertise and narrative, as complex cognitive capacities, have each been investigated quite deeply, but little attention has been paid to their interdependence. We offer here the position that treating these two domains together can fruitfully inform the modeling of expert cognition and behavior, and present the framework we have been using to develop this approach, the SGOMS macro-cognitive architecture. Finally, we briefly explore the role of narrative in an SGOMS model of cooperative video game playing.
news summary (7)

Weekly QuEST Discussion Topics and News, 15 Aug

August 14, 2014 Leave a comment

Weekly QuEST Discussion Topics

 15 Aug 2014

The first topic is an article from the news stories this week – ‘Forget Siri: this Radical New AI Teaches itself and reads your mind’ – by Steven Levy, it discusses Viv which is an attempt to overcome the limitations of digital assistant software associated to only being able to respond / perform tasks that have specifically been implemented by the designing engineers.  Viv does this by teaching itself and capturing the new ‘thoughts’ by writing its own new code.  It also captures a representation of a particular user so it can generate anticipatory actions.  QuEST needs to discuss this work from several angles.  The limitations of current solutions helps us refine our thoughts on ‘unexpected queries’.  We want to look at some of the examples provided and discuss how a conscious agent solves them and how an artificially conscious QuEST agent could hope to bring value.


Forget Siri: This Radical New AI Teaches Itself and Reads Your Mind


The next topic is a brief discussion about a short article provided by our colleague Robert P.  ‘Thinking about thinking’.  It stimulated an exchange where we revisited a prior tenet discussion where Prof Oxley pointed out to us that there had been a mathematical proof we needed to consider about self referential systems – to resulting discussion led to our addition of the following tenet:


Thinking about thinking:  A mathematical proof has been constructed that theorizes it is impossible to think about yourself thinking.  This is consistent with the ‘One quale at a time’ tenet which suggests that you can only think about having thought.  You can only be conscious of having been conscious.  You are never conscious of being conscious.


This brings up the question of accounting for meta-cognition.  We might want to have this discussion.

Next there was a request for publications from our group from a journal Philosophy Study.  While looking through the journal we noticed an article by Deepak Chopra: ‘From Quanta to Qualia: How a paradigm Shift Turns Into Science’.  Our colleague Mike Y and Capt Amerika have read through the article and we might spend a couple minutes talking about the area of Panpsychism and how QuEST would respond.

Lastly if there is time there are several articles from the computational models of narratives conference that caught our eye.  We might go through a couple of them if they turn out to be interesting from a QuEST implementation perspective.  The first is Modeling the Function of Narrative in Expertise by MacDougall et al.

Abstract: The use of narrative is ubiquitous in the development, exercise, and communication of expertise.  Expertise and narrative, as complex cognitive capacities, have each been investigated quite deeply,

but little attention has been paid to their interdependence. We offer here the position that treating these two domains together can fruitfully inform the modeling of expert cognition and behavior, and present the framework we have been using to develop this approach, the SGOMS macro-cognitive architecture. Finally, we briefly explore the role of narrative in an SGOMS model of cooperative video game playing.

news summary (5)

Categories: Uncategorized

CSE/EID Workshop

August 14, 2014 Leave a comment

See below here for details on a workshop on Cognitive Systems Engineering (CSE) and Ecological Interface Design (EID), as presented by Dr. John Flach and Dr. Kevin Bennett.

Workshop Supporting Productive Thinking through Interface Design

This workshop is designed to introduce researchers and practitioners to the Cognitive
Systems Engineering (CSE) paradigm and the implications for the design of interfaces
to support problem solving in complex work domains (Ecological Interface Design –
EID). Basic assumptions about the pragmatic nature of reasoning (i.e., problem solving
and decision making) in complex work domains will be presented. The CSE/EID
framework provides a holistic systems perspective for addressing a wide range of
issues associated with performance in sociotechnical systems (e.g., workload,
situation awareness, trust, distributed collaboration, resilience, etc.). Particular
attention will be paid to the implications for the design of interface representations
that enhance perspicacity with respect to the deep structure of work domains in order
to support productive thinking.

Where: Building 441 Auditorium
When: Wednesday, August 20, 2014; 0900-1230
Kevin Bennett and John Flach recently co-authored the text Display and Interface Design.
This text is the culmination of a nearly 25 year collaboration to explore Cognitive Systems
Engineering as an alternative approach for leveraging human capabilities to achieve skilled
control in complex work environments. Their approach has been praised for the ability to
link basic theories from ecological psychology/situated cognition to innovative design

Categories: Uncategorized

Proceedings from Computational Model of Narrative 2014

August 11, 2014 Leave a comment

From Sandy Vaughn, below please find the url link to the proceedings, as well as some papers with potential application to QuEST efforts.

The papers which in my opinion apply most to the QuEST research area (don’t judge these papers applicability by their titles):

Mindreading, Privileged Access and Understanding Narratives
Authors: Kiss, Szabolcs ; Jakab, Zoltán
(“…how children develop their internal subjective representations (qualia)” – he refers to Wittgenstein and uses ‘qualia’ in his paper in the same sense that we are.)

Toward Neurally-Inspired Computational Models of Narrative (Invited Talk)
Authors: Zacks, Jeffrey M.

Modeling the Function of Narrative in Expertise
Authors: MacDougall, W. Korey ; West, Robert L. ; Genovesi, Christopher
(modeling in ACT-R, and interesting reference to Newell’s task levels)

A Hybrid Representational Proposal for Narrative Concepts: A Case Study on Character Roles
Authors: Lieto, Antonio ; Damiano, Rossana
(Wittgenstein – family resemblance, Prototype Theory and )

A Character Model with Moral Emotions: Preliminary Evaluation
Authors: Battaglino, Cristina ; Damiano, Rossana
(Discusses Affect, Dual-process (Stanovich), derivation model, and Intensity model of emotions, reasoning with exceptions, representation of common sense)

Weekly QuEST Discussion Topics and News, 8 Aug

August 8, 2014 Leave a comment

The first topic this week is to allow our colleague Sandy V to provide us an after action report on her presentation / interactions at the Computational Models of Narratives conference – recall her paper:

Abstract for ‘Narratives as a Fundamental Component of Consciousness’

In this paper, we propose a conceptual architecture that models human (spatially-temporally-modally) cohesive narrative development using a computer representation of quale properties. Qualia are proposed to be the fundamental cognitive components humans use to generate cohesive narratives. The engineering approach is based on cognitively inspired technologies and incorporates the novel concept of quale representation for computation of primitive cognitive components of narrative. The ultimate objective of this research is to develop an architecture that emulates the human ability to generate cohesive narratives with incomplete or perturbated information.

The next topic is a couple of articles provided to us by our colleague Robert P. – specifically last week we reviewed the ‘mind’s best trick’ article – we were running out of time before we could have an in depth discussion on the implications to our QuEST framework – do we want deliberations with the aspects of our representation that we are conscious of to be the cause of actions or is this just an illusion – we want to have a discussion on the implications of these alternatives on our QuEST implementations

The mind’s best trick:
how we experience conscious will
Daniel M. Wegner

• We often consciously will our own actions. This experience is so profound that it tempts us to believe that our actions are caused by consciousness.
• It could also be a trick, however – the mind’s way of estimating its own apparent authorship by drawing causal inferences about relationships between thoughts and actions.
• Cognitive, social, and neuropsychological studies of apparent mental causation suggest that experiences of conscious will frequently depart from actual causal processes and so might not reflect direct perceptions of conscious thought causing action.

Before we go to the second article – the discussion above is related to another topic that came up this week while working with our colleagues Bob Mills / John R. / Dean W. – specifically we were discussing cybernetics (interactions feedback and intelligence) and that reminded us of a think piece we had written ‘Beyond the OODA nonsense’. We will revisit the issues we attempted to address in that article from the perspectives of the discussion above on conscious will causing action.

Then the second article sent to us by Robert P

Johansson, P., Hall, L., Sikström, S., & Olsson, A. (2005). Failure to detect
mismatches between intention and outcome in a simple decision task.
Science (New York, N.Y.), 310(5745), 116–9. doi:10.1126/science.1111709

• A fundamental assumption of theories of decision-making is that we detect mismatches between intention and outcome, adjust our behavior in the face of error, and adapt to changing circumstances. Is this always the case? We investigated the relation between intention, choice, and introspection. Participants made choices between presented face pairs on the basis of attractiveness, while we covertly manipulated the relationship between choice and outcome that they experienced. Participants failed to notice conspicuous mismatches between their intended choice and the outcome they were presented with, while nevertheless offering introspectively derived reasons for why they chose the way they did. We call this effect choice blindness.