Archive

Archive for December, 2012

Weekly QUEST Discussion Topics and News, Dec 21st

December 20, 2012 Leave a comment

This will be the last QUEST meeting of 2012. Meetings will resume after the holiday break on Jan 4th, 2013.

QUEST Discussion Topics and News
December 21st, 2012

Our systems health monitoring colleagues have requested that we revisit a topic that they’ve been pursuing – the framework article: The Function of Phenomenal States: Supramodular Interaction Theory by Ezequiel Morsella of Yale University, the article was provided to us by our colleague Mike Young. Discovering the function of phenomenal states remains a formidable scientific challenge… integrate diverse kinds of information to yield adaptive action. Supramodular interaction theory (SIT )proposes that phenomenal states play an essential role in permitting interactions among supramodular response systems… Without phenomenal states, these systems would be encapsulated and incapable of collectively influencing skeletomotor action

This leads to a prior topic – Summary notes on designing conscious systems – goal is defining ‘What makes a quest solution v6’ – A discussion that mixes architecture / function to sneak up on what would be a QUEST solution.

QUEST Discussion Topics and News Dec 21st

Advertisements

Weekly QUEST Discussion Topics and News, Dec 14th

December 13, 2012 Leave a comment

Weekly QUEST Discussion Topics and News Dec 14th

Weekly QUEST Discussion Topics and News
December 14, 2012

Making the hard problem easy – a Theory of Consciousness
This week we want to hit the article by David Chalmers – ‘The puzzle of conscious experience’ and specifically map the discussion to our efforts to define the engineering characteristics of conscious experience – thus making the hard problem (Why does it exist? What does it do? How could it possibly arise from neural processes in the brain?) easy. I put in bold the questions I think we have been addressing versus the neural correlate questions. Chalmers suggests – consciousness might be explained by a new kind of theory. The full details of such a theory are still out of reach, but careful reasoning and some educated inferences can reveal something of its general nature. For example, it will probably involve new fundamental laws (*** these are the QUEST tenets we’ve listed ***),and the concept of information may play a central role (** we certainly early on in our efforts defined the unambiguous parsing of stimuli, data and information concluding a subjective definition was required **). These faint glimmerings suggest that a theory of consciousness may have startling consequences for our view of the universe and of ourselves (** and we would add have a dramatic impact on our engineering solutions for machine-human collaboration ***).
These are questions that we would like a theory of consciousness to answer.
• Why do we have any experience at all? There is an engineering advantage to an approach that uses the representational characteristics of our illusory Cartesean theater = qualia.
• Could not an unconscious automaton have performed the same tasks just as well? My position is the answer to this question is NO – to achieve the robustness that nature demonstrates requires the ‘conscious’ representation.
• Why do we experience anything at all? Engineering advantage.
• What leads to a particular conscious experience (such as the blueness of blue)? Subjectively unique, innate initial condition modulated by experience.
• Why are some aspects of subjective experience impossible to convey to other people (in other words, why are they private)? Situated nature allows comparison with other conscious experience but they don’t exist in isolation.
An approach to the hard questions should be based around ‘Meaning’ – implies a connection to ‘information’ – not Shannon info but information as meaning – Meaning derives from the linkages – the interesting question becomes what engineering characteristics must be present to capture the essence of meaning achieved by ‘consciousness’ in nature – although Chalmers takes the position – Conscious experience as a fundamental feature = irreducible – we are suggesting there are fundamental laws of the representation from which we can derive characteristics and predict performance – we seek fundamental laws : start by looking for high level bridging laws, connecting interacting processes to account for experience at an everyday level –

QUEST Discussion Topics and News Dec 7th

December 7, 2012 Leave a comment

QUEST Discussion Topics and News Dec 7th

QUEST Discussion Topics and News
7 Dec 2012:

We have 4 topics that have dominated our interactions this week to talk about:

1.) We want to discuss ideas of an experiment that can distinguish between what an analyst is ‘conscious’ of in a data set versus what they gazed at while looking at the data.

The application requires the human to call out stuff (like we are doing in dgs) and assume we have access to speech processing entry into the products generated and we have the means to determine the gaze pattern (gaze does NOT equal conscious attention) of the analyst while reviewing the image/video data – just because they don’t call it out doesn’t mean they are not conscious of it – but if they never even gazed in the direction of a given stimuli AND they didn’t call it out we can conclude they were NOT conscious of it – thus the point – a tool (like an object recognition system) that has this information (the information about what the human analyst was directed to attend to AND information about what we can deduce they were conscious of and not conscious of) and can thus change its ‘weighting’ of potential areas of concern could be a better partnering agent.

2.) The second topic is similarly a discussion of an experiment / research concentration for our AFRL researcher hostage exchange with Google. To have this discussion I want to bring up at least in a cursory manner a think piece we’ve been developing on ‘Google becoming Conscious’. This is a derivative of the interview with Koch on the internet becoming conscious. The idea of the think piece is that the infrastructure Google has created in its file system has many of the characteristics we’ve been defining as the engineering characteristics of the sys2 processing that culminates in the Qualia representation. The idea of this discussion is to help refine ideas that our researchers can pursue when provided access to the infrastructure of Google and investigate whether the addition of the ‘situationally conscious’ layer on top of the current reflexive sys1 Google infrastructure can provide some engineering advantage for task in large data processing.

3.) The next topic is an article 2003 by Crick and Koch on a Framework for Consciousness – the framework is focused on visual consciousness but is quite general.

4.) The last topic is also a framework article: The Function of Phenomenal States: Supramodular Interaction Theory by Ezequiel Morsella of Yale University, the article was provided to us by our colleague Robert Patterson. Discovering the function of phenomenal states remains a formidable scientific challenge… integrate diverse kinds of information to yield adaptive action. Supramodular interaction theoryproposes that phenomenal states play an essential role in permitting interactions amongsupramodular response systems… Without phenomenal states, these systems would be encapsulated and incapable of collectively influencing skeletomotor action