Home > Uncategorized > QUEST Discussion Topics, Sept 30

QUEST Discussion Topics, Sept 30

1.) Topic: Capt Amerika, adam and ox had a conference call with the Affectiva group on Friday and will provide walk-away points on their technology and potential touch points for our interests.
2.) Topic: a recent news article (24 sept 2011 The Daily) http://kendrickkay.net/KayNature2008.pdf
Identifying natural images from human brain activity, Kay, Naselaris, Prenger, Gallant, vol 452 20 march 2008 nature) on ‘seeing what you are thinking’ – the idea of a general purpose visual decoder – generating a visual representation of what you are seeing so someone else can view it – ‘reconstruct a picture of a person’s visual experience at any moment in time’ – I have real issues with this last statement so I want to review what I think they are doing and how it relates to our QUEST model of sys1 and sys2.
3.) Topic: A recent article on a ‘theory of meaning’ http://homepages.feis.herts.ac.uk/~dc09aav/publications/context-distributional-semantics.pdf
A context-theoretic framework for compositionality in distributional semantics, Clarke, 2005 Association for Computational Linguistics)– although only a framework it does provide a path to extend the work of our prior QUEST colleague Bobby Birrer (meaning based on statistical analysis) into a ‘context theory’ for meaning – implying meaning is all in the context (meaning of an expression is determined by how it is used) – although it will take the QUEST mathematicians to explain the math we at least want to discuss why we need a theory of meaning – this work is focused on the text application but we’ve said with respect to qualia – ‘by their links you shall know them’ – trying to capture the idea that qualia can only be described to others by communicating the relationship to other qualia (we’ve sometimes termed this the dictionary problem – you define a word in terms of other words) – since words are used to communicate qualia the fact that these issues are related is not surprising.
4.) Topic: to continue the brainstorming session on the BURKA lab experiment and its potential use of their modeling / simulation environment as the sys2 cognition engine for a quest agent. This week we provided to some a homework assignment – redraw the conventional single / multi-sensor ATR diagrams using the concepts we’ve discussed in QUEST. We want to review the ideas of the audience on how we would propose to revolutionize one of our core missions (ATR – automatic target recognition). We will start by reviewing the current key ideas to improve ATR. Then we will discuss ideas on mapping those into our QUEST intuition / deliberative dual system formalism. Our hope is that this will lead to a specific set of ideas for the Burka lab researchers (for example how to improve tracking …).
5.) We’ve proposed one Burka lab experiment to investigate the application of QUEST concepts to solve the problem of what bits to exploit and at what resolution in a layered sensing environment. The idea is that we don’t have the option of processing all bits sensed for all possible meanings so we need an approach to decide what we look at and how we look at it. The best analogy is snail mail. When we go to the mail box we sort the material at different levels of resolution. Some things we discard by just looking at the envelope. Some we read the details of the source then discard. Some we actually have to open to discard. Some we have to read all the contents to decide how to process. The idea is to architect a multi-resolution Burka experiment where conflicts between the ‘simulation’ (sys2 representation) and the observations at coarse resolutions can drive exploitation resources and when those conflicts don’t exists we can avoid expending those exploitation resources and potentially not drown in that data. We specifically want to add the details for ATRs and Trackers. We would also like to discuss the potential integration of human analyst in this discussion using technology to measure their sys1 response to a given set of data at a given resolution. We will examine the devices offered by the company Affectiva for this purpose.
6.) Topic Affectiva: a continuation of our discussion – ‘can you make objective measurements of emotions or pain or even ‘redness’?’ – related to our cloud diagram that we were using as a homework assignment. Specifically accounting for relationships between external observables and aspects (both sys1 {subconscious} and sys2 {conscious}) of our internal representation of the world is our goal.
a. For example, when we say the word ‘red’ to describe some spectral aspect of some part of our visual sensory data that we’ve incorporated into our illusory cartesean theater what is the source of that articulated label and what is its relationship to the visual quale {the actual experience of what is seen} and to the sensory data {what is captured at the retina and encoded in pulses}.
b. We want to have the same discussion with respect to more complicated qualia like emotions and pain. We would like to account for the relationships between ‘frustration’ or ‘confusion’ to external observables like electro-dermal activity (EDA) and the internal qualia and also the internal sys1 states that we don’t have available to introspect over.
c. The focus of the discussion is to establish a position on how we can design an engineering experiment to allow quest agents to capture an accurate representation of the humans they are collaborating with in order to better estimate what context to provide that human to improve decision making.
7.) Topic: a recent article on impact of processing of scenes.
http://www.sciencenews.org/view/generic/id/334047/title/If_that%E2%80%99s_a_TV%2C_this_must_be_the_den
Specifically the impact of object recognition integrated within the context of the general outline of the scene information. The researchers presented to 28 people four scenes (bathroom, kitchen, intersection, and a playground). They then were presented objects associated with entities in those environments and the neural signatures for those objects representations were recorded (specifically in the lateral occipital cortex – LOC). The combination of the simple object representations was then compared to the scene presentation. The combination of the stove and the fridge responses matched the response to the kitchen. The implication is that within the LOC the representation of the scene is a simple combination of parts.
8.) Someday we will get to the other topic – the recent Sci Amer ‘Mind’ issue July 2011, a word doc with some snippets from some of the articles can be provided to stimulate discussion.

Advertisements
Categories: Uncategorized
  1. No comments yet.
  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: