Sunday, January 29, 2012
Berkeley talk: online consciousness course begins Feb 7 2012
“Consciousness is cheap; organismal metabolism and evolutionary constraints on brain size”
Sean O Nuallain PhD, Room 28, Ventura Hall, Stanford
Whaley lab group meeting, UC Berkeley
Location; top floor, 410 Hearst Memorial Mining Building, UC Berkeley
Date/Time; Wed Feb 1, noon.
Intelligence requires bigger brains. Such brains are metabolically expensive, and make childbirth painful and dangerous for us bipedal animals. It may be the case that humans have reached the limit for brain size, that we have compressed to the furthest point that ion channels are still reliable by keeping neurons small
Consciousness seems to have an ambiguous role in our functioning. In particular, classic work by Libet suggests that motor neurons show preparation to initiate action tenths of seconds before we are aware that we intend to do so It is not a stretch to propose that we continually narrate to ourselves a “stream of consciousness” about our actions that is little more than what a sports commentator does about a game he is calling
However, it does seem to be the case that consciousness coincides with a episode of synchronized gamma, and that such episodes betoken a decrease by 4 orders of magnitude in the metabolic demand by the brain on the rest of the organism. It is now accepted that the brain consumes around 20% of the organism's total metabolic energy. That leads to a hypothesis; perhaps consciousness is fundamentally a “spandrel”, an incidental artifact of the need to keep metabolic costs down
In this talk, the relevant evidence – phenomenological and evolutionary – will be considered, and the state of knowledge about mechanisms of consciousness will be outlined. We will then proceed to examine the justification for what seems a priori a counter-intuitive hypothesis
Relevant publication
http://www.springerlink.com/content/x10063878485504n/
Speaker Bio
Sean O Nuallain holds a Ph.D. in Computer Science from Trinity College, Dublin. He is the author and editor of 5 books in cognitive science and Consciousness studies, with "The Search for Mind" running to three editions. From 2005, he has focused on experimental neuroscience and theoretical biology, and has held visiting positions teaching and researching at both UC Berkeley and Stanford while publishing in these areas.
From Feb 7 2012 until April 8 2012 he is offering the course “Neuroscience and experience” precisely as taught as an advanced seminar in Stanford. A sample lecture and outline of the course can be found at
http://floyddogdesign.com/sean/newsyll2010.html
Queries can be sent to
universityofireland@gmail.com
It does NOT use video, as we believe that slides + voiceover is a more economical means of learning. The method of assessment is essay submission at the end of the course. As with all the other courses recently offered following the AI model, it is NOT accredited by Stanford, but taught exactly as in Stanford
Seán O Nualláin Ph.D. Stanford
Stanford University 29u Eanair 2012
Sean O Nuallain PhD, Room 28, Ventura Hall, Stanford
Whaley lab group meeting, UC Berkeley
Location; top floor, 410 Hearst Memorial Mining Building, UC Berkeley
Date/Time; Wed Feb 1, noon.
Intelligence requires bigger brains. Such brains are metabolically expensive, and make childbirth painful and dangerous for us bipedal animals. It may be the case that humans have reached the limit for brain size, that we have compressed to the furthest point that ion channels are still reliable by keeping neurons small
Consciousness seems to have an ambiguous role in our functioning. In particular, classic work by Libet suggests that motor neurons show preparation to initiate action tenths of seconds before we are aware that we intend to do so It is not a stretch to propose that we continually narrate to ourselves a “stream of consciousness” about our actions that is little more than what a sports commentator does about a game he is calling
However, it does seem to be the case that consciousness coincides with a episode of synchronized gamma, and that such episodes betoken a decrease by 4 orders of magnitude in the metabolic demand by the brain on the rest of the organism. It is now accepted that the brain consumes around 20% of the organism's total metabolic energy. That leads to a hypothesis; perhaps consciousness is fundamentally a “spandrel”, an incidental artifact of the need to keep metabolic costs down
In this talk, the relevant evidence – phenomenological and evolutionary – will be considered, and the state of knowledge about mechanisms of consciousness will be outlined. We will then proceed to examine the justification for what seems a priori a counter-intuitive hypothesis
Relevant publication
http://www.springerlink.com/content/x10063878485504n/
Speaker Bio
Sean O Nuallain holds a Ph.D. in Computer Science from Trinity College, Dublin. He is the author and editor of 5 books in cognitive science and Consciousness studies, with "The Search for Mind" running to three editions. From 2005, he has focused on experimental neuroscience and theoretical biology, and has held visiting positions teaching and researching at both UC Berkeley and Stanford while publishing in these areas.
From Feb 7 2012 until April 8 2012 he is offering the course “Neuroscience and experience” precisely as taught as an advanced seminar in Stanford. A sample lecture and outline of the course can be found at
http://floyddogdesign.com/sean/newsyll2010.html
Queries can be sent to
universityofireland@gmail.com
It does NOT use video, as we believe that slides + voiceover is a more economical means of learning. The method of assessment is essay submission at the end of the course. As with all the other courses recently offered following the AI model, it is NOT accredited by Stanford, but taught exactly as in Stanford
Seán O Nualláin Ph.D. Stanford
Stanford University 29u Eanair 2012
Sunday, January 22, 2012
How many universities is too many?
Readers with little time on their hands can simply read this first
sentence take-home message; all the Irish universities should, as
matter of urgency, be rolled into a revamped National university, and
all degrees awarded to date by DCU and UL should be validated by that
body. This protects the 100k or so graduates of UL and DCU, while
acknowledging that both institutions were given far too much autonomy,
and university status way too early. Now, of course, they are rushing to
pull up the ladder after them, and deny university status to Waterford/
Carlow;
http://www.irishtimes.com/newspaper/ireland/2012/0123/1224310627145.html
Direct link
There is also a strong case for selling off both UL and DCU, and indeed many of the state-owned “third level” sites in the country. The reason; while this unseemly row is going on, the provisional tallies for Stanford's upcoming course enrollment has just been announced; the new diet of free courses has an average enrollment of over 40k, and MIT is about to enter the fray. By years' end, there will be over a million enrolled at both Stanford and MIT; in fact, Stanford has just about hit the million mark already with 17 courses in 2011-2012. There is no reason to try and put forward any competition; in this case, resistance indeed is futile. Full disclosure; I have worked at Stanford since 2002.
The smart thing for the Irish government to do would be to get ahead of the curve, and acknowledge that tertiary education changed irrevocably once Norvig, Thrun and Ng of Stanford demonstrated that classes of hundred of thousands in engineering subjects were technologically feasible. The minor cavil offered – that correction of computer programming assignments is coarse grained – simply amplifies the critical point that, desperate for ranking points, Ireland moved away a decade ago from TEACHING at its universities. Specifically, resources hitherto put into hiring tutors to help students learn skills like programming were redeployed into "research".
In fact, as repeatedly pointed out, Ireland's universities were to be a neoliberal experiment in unfettered use of state power to pursue a crony capitalist agenda. the levels of abuse of staff and students allowed are what motivated this blog. The Stanford/MITX initiatives put power back where it should be; in the hands of scholars, teachers and students alike.
Oh, the rankings! In fact, none of the Irish universities are in the top 200 in the only one that should matter – the shanghai. DCU is not even in the top 500. Why? Essentially, because the Chinese know that the number of Nobel laureates happy to work at a university is a better index to quality than the percentage of foreign students and foreign lecturers. Consider this; the Irish taxpayer is asked to pay for a very expensive system which bows to a rating system that values foreign scholars over ones genetically related to that taxpayer. The people for whom this charade is presumably created – the Chinese – are not in the slightest bit impressed.
Anyway, grab a bag of popcorn and enjoy the show as the 7 presidents who refused to cut their pay to 200k battle it out with the Labour party, The many sincere people in the Irish provinces who want a college should look at how the world is changing and adapt
Seán O Nualláin Ph.D. Stanford
Stanford University 22u Eanair 2012
PS 25 Ean 2012; Two recent events are probably not unrelated -
1. Those of us auditing some of the new Stanford mass courses (In my case, NLP) were told the courses were delayed;
2. Sebastian Thrun announced he was resigning from his tenured position at Stanford to set up a private university
udacity.com
outline syllabi
Udacity lays claim to ownership of the AI course, as you find in the intricate legal document you need to assent to get access to the "free" courses.
The connection? I went to the Comp Sci dept at Stanford last week to be told they had NOTHING to do with the mass courses.
PPS 18 Feabhra 2012 All is now clear; Ng and Koller have set up a company Coursera which will run the great majority of these courses once Vice-President Mitchell approves them, MITX starts Mar 5 2012, and tertiary education is finally undergoing a revolution greater and more beneficial than that which happened at Gutenberg, a revolution in which the entire world can participate.
This morning I got the following, confirming my view that there has been a row;
"We're sorry to have to tell you that our Natural Language Processing course will be delayed further. There have naturally been legal and administrative issues to be sorted out in setting up an online course infrastructure that works both for Stanford and the outside world, and it's just been taking time. We have however been able to take advantage of the extra time to debug and improve our course content!
We now expect that the course will start either late in February or early in March."
In fact, Coursera is not even being allowed to deliver courses already prepared - and seems to have moved material from Youtube
Some readers will know that in 1989 I set up the first specialist course - in fact, a whole degree program - in Ireland in Natural Language Processing. It was highly successful while I ran it, but was scrapped in 2003 - a DCU insider told me in order to foment redundancies - and millions of taxpayer $ were put into garbage projects like a "localization" center instead. So we have lost a generation.
26 u Feabhra 2012 - further corrections above after Daphne and Andrew told me that Mitchell is holding things up at Stanford, even for courses ready to go. Youtube seems to have lost some material; it is now on the Coursera website, which means no longer necessarily Creative commons/Youtube licensed
http://www.irishtimes.com/newspaper/ireland/2012/0123/1224310627145.html
Direct link
There is also a strong case for selling off both UL and DCU, and indeed many of the state-owned “third level” sites in the country. The reason; while this unseemly row is going on, the provisional tallies for Stanford's upcoming course enrollment has just been announced; the new diet of free courses has an average enrollment of over 40k, and MIT is about to enter the fray. By years' end, there will be over a million enrolled at both Stanford and MIT; in fact, Stanford has just about hit the million mark already with 17 courses in 2011-2012. There is no reason to try and put forward any competition; in this case, resistance indeed is futile. Full disclosure; I have worked at Stanford since 2002.
The smart thing for the Irish government to do would be to get ahead of the curve, and acknowledge that tertiary education changed irrevocably once Norvig, Thrun and Ng of Stanford demonstrated that classes of hundred of thousands in engineering subjects were technologically feasible. The minor cavil offered – that correction of computer programming assignments is coarse grained – simply amplifies the critical point that, desperate for ranking points, Ireland moved away a decade ago from TEACHING at its universities. Specifically, resources hitherto put into hiring tutors to help students learn skills like programming were redeployed into "research".
In fact, as repeatedly pointed out, Ireland's universities were to be a neoliberal experiment in unfettered use of state power to pursue a crony capitalist agenda. the levels of abuse of staff and students allowed are what motivated this blog. The Stanford/MITX initiatives put power back where it should be; in the hands of scholars, teachers and students alike.
Oh, the rankings! In fact, none of the Irish universities are in the top 200 in the only one that should matter – the shanghai. DCU is not even in the top 500. Why? Essentially, because the Chinese know that the number of Nobel laureates happy to work at a university is a better index to quality than the percentage of foreign students and foreign lecturers. Consider this; the Irish taxpayer is asked to pay for a very expensive system which bows to a rating system that values foreign scholars over ones genetically related to that taxpayer. The people for whom this charade is presumably created – the Chinese – are not in the slightest bit impressed.
Anyway, grab a bag of popcorn and enjoy the show as the 7 presidents who refused to cut their pay to 200k battle it out with the Labour party, The many sincere people in the Irish provinces who want a college should look at how the world is changing and adapt
Seán O Nualláin Ph.D. Stanford
Stanford University 22u Eanair 2012
PS 25 Ean 2012; Two recent events are probably not unrelated -
1. Those of us auditing some of the new Stanford mass courses (In my case, NLP) were told the courses were delayed;
2. Sebastian Thrun announced he was resigning from his tenured position at Stanford to set up a private university
udacity.com
outline syllabi
Udacity lays claim to ownership of the AI course, as you find in the intricate legal document you need to assent to get access to the "free" courses.
The connection? I went to the Comp Sci dept at Stanford last week to be told they had NOTHING to do with the mass courses.
PPS 18 Feabhra 2012 All is now clear; Ng and Koller have set up a company Coursera which will run the great majority of these courses once Vice-President Mitchell approves them, MITX starts Mar 5 2012, and tertiary education is finally undergoing a revolution greater and more beneficial than that which happened at Gutenberg, a revolution in which the entire world can participate.
This morning I got the following, confirming my view that there has been a row;
"We're sorry to have to tell you that our Natural Language Processing course will be delayed further. There have naturally been legal and administrative issues to be sorted out in setting up an online course infrastructure that works both for Stanford and the outside world, and it's just been taking time. We have however been able to take advantage of the extra time to debug and improve our course content!
We now expect that the course will start either late in February or early in March."
In fact, Coursera is not even being allowed to deliver courses already prepared - and seems to have moved material from Youtube
Some readers will know that in 1989 I set up the first specialist course - in fact, a whole degree program - in Ireland in Natural Language Processing. It was highly successful while I ran it, but was scrapped in 2003 - a DCU insider told me in order to foment redundancies - and millions of taxpayer $ were put into garbage projects like a "localization" center instead. So we have lost a generation.
26 u Feabhra 2012 - further corrections above after Daphne and Andrew told me that Mitchell is holding things up at Stanford, even for courses ready to go. Youtube seems to have lost some material; it is now on the Coursera website, which means no longer necessarily Creative commons/Youtube licensed
Tuesday, January 10, 2012
The "zero power and selflessness" paper
I published this some years ago; it has been referenced a lot, and cited
a few times. Its publishers have failed to make it available on the
web, and the contract I signed with them allows me to post the following
selected pages;
Seán O Nualláin Ph.D. Stanford
Stanford University 11-12u Eanair 2012
PS 15 u Eanair 2012
Much ink and indeed many petabytes have been consumed in an attempt to work out physical mechanisms for quantum coherent states in biology. Ironically, this reached a crescendo in consciousness studies in the 1990's - as if we didn't have enough problems of our own! It was hypothesised that consciousness involved microtubules, or perhaps laying down of dendritic spines in a way reminiscent of quasiperiodic crystals - even before Steinhardt found the first attested quasiperiodic crystal in nature in 2011. Clearly, phase-synchronized gamma would be conducive to coherence
Maybe it's simpler. Maybe all we need is coherence due to entanglement as is hpothesised for birds' magnetic navigation - and entanglement has been demonstrated for relatively macroscopic objects. The framework of the paper above still remains intact. I will give a paper on this and other issues Feb 1 2012 at UC Berkeley
Seán O Nualláin Ph.D. Stanford
Stanford University 15u Eanair 2012
Seán O Nualláin Ph.D. Stanford
Stanford University 11-12u Eanair 2012
PS 15 u Eanair 2012
Much ink and indeed many petabytes have been consumed in an attempt to work out physical mechanisms for quantum coherent states in biology. Ironically, this reached a crescendo in consciousness studies in the 1990's - as if we didn't have enough problems of our own! It was hypothesised that consciousness involved microtubules, or perhaps laying down of dendritic spines in a way reminiscent of quasiperiodic crystals - even before Steinhardt found the first attested quasiperiodic crystal in nature in 2011. Clearly, phase-synchronized gamma would be conducive to coherence
Maybe it's simpler. Maybe all we need is coherence due to entanglement as is hpothesised for birds' magnetic navigation - and entanglement has been demonstrated for relatively macroscopic objects. The framework of the paper above still remains intact. I will give a paper on this and other issues Feb 1 2012 at UC Berkeley
Seán O Nualláin Ph.D. Stanford
Stanford University 15u Eanair 2012
Thursday, January 5, 2012
So what was all that about, then?
The 1997 act creates a “chief officer' at the Irish universities in whom all aspects of the universities converge. Whether such a role has any place in a democratic society is a moot point; yet a succession of Irish “chief officers”, egged on by the government, acted outside the law for over a decade, with dreadful consequences for people's lives. Now that the neoliberal dispensation of which this was apart has failed, there will be hell to pay for this violation of the social contract
This writer must confess that he oscillates between anger at the Irish Times for creating VP as a media figure, and relief. The relief arises because this cast VP out with the other bullshitters who expostulate on our national media. Unlike the rest of them, VP had actual power; yet he failed to use it, and we are all the better for this. However, there is a more fundamental point.
Academic freedom will henceforth primarily be a function of how the state chooses to police the internet. The Irish tradition is MacSwiney's; intellectual freedom is about our choice in our lives. The main significance of what went on at the Irish universities 1999-2011 is the attempt to create an entity, funded by te taxpayer, that could function outside the law. It is in many ways incidental that it happened at the universities; something similar was going on in the health service.
The effects on students and staff has been documented here. Thanks to some brilliant academics at Stanford (all their course materials are available free) we now have the capacity to say “Nie wieder!” (never again!). Accreditation is now a nonsense; 20,000 students passed the AI course, 6,100 + passes teh databases course and now try and tell your employer that your “accredited' course is better than the Stanford one. What I conceive of is a transfer of credits scenario, and here is how it would work out for the cog sci course.
_____________________________________________________________
A cognitive science course, reparsed
Stanford courses as credits (coded as *) – we assume an average of 4 credits/units each. 45 credits must be gained, and this will be done over 3-4 years
A sign-off by the lecturer or audited work during course as assessment. It is not insinuated that these are accredited by Stanford. Rather, what is happening is that certain of these courses, offered by Stanford faculty on-line, are acceptable to us as equivalents of our courses if the lecturer at Stanford signs a certificate of satisfactory completion for the student. Alternatively, the student can take only our in-house courses. Stanford courses are marked *; my own, already available online, **
Introductory material – for example, linear algebra – is often reduplicated over several courses (eg ML and AI) and in any case there are many good and free tutorials on the web For several of the courses, my “search for Mind” is used
Year 1
Semester 1
Algorithms * http://www.algo-class.org/
CS 101 ( may be an essential for the computing stream) *http://www.cs101-class.org/
Cog sci ; chapters 1 and 2 of “search”
Semester 2
Software as a service (Berkeley) * http://www.saas-class.org/
Sci and Society – 2 credits **
Cog sci Chapter 3 and 4 of “search”
Irish culture - 2 credits **
Year 2
Semester 1
Databases (encompasses bioinformatics) *http://www.db-class.org/course/class/index
Biosemiotics **
Intro to AI (encompasses some of biosemiotics) *https://www.ai-class.com/
Semester 2
Complex systems *http://www.modelthinker-class.org/
Nat lang proc (encompasses some of biosemiotics also) * http://www.nlp-class.org/
Cog sci; Remainder of “search”
HCI *http://www.hci-class.org/
Year 3
ML * http://www.ml-class.org/course/class/index
Neuroscience and experience **
Specialist projects; work experience; on-line discussions. Likewise for year 4 if this is necessary
Seán O Nualláin Ph.D. Stanford
Stanford University 5u Eanair 2012
