Cognitive Computing Consciousness and Computation Advanced topics in

Cognitive Computing Consciousness and Computation Advanced topics in Cognitive Computation Lent term assignments Mark Bishop

Consciousness and Computations • Formative (30%) – Prove that the machine NORMA_STACK is no more powerful than the universal machine NORMA by designing two MACROs to implement: • (a) X=POP which removes the top value from the STACK and places it into the X register; (b) PUSH (X) which places the contents of the X register on to the top of the STACK; • and submit a short (no more than 1 -page) report detailing their operation. • Summative (70%) – A 3000 word researched essay, “An introduction to Wiedermann and Leeuwen’s ‘Autopoietic Automata’ (AA) theory” (submitted by 5 pm, Friday 22 nd March, 2013). • It is anticipated that – at minimum - student essays will address: (a) the Turing model of computation; (b) the Interaction model of computation (c) what is an AA? (d) what is the relation of AAs to other models of computation (e. g. Turing machines, Interaction machines etc. ); – (e) what is the computational power of an AA; – (f) the implications [if any] of AA (and more generally the interactive model) for cognitive science. – –

References • Amongst other sources students are recommended to consult: – Jan van Leeuwen & Jirı Wiedermann (2001): Beyond the Turing Limit: Evolving Interactive Systems. In: Leszek Pacholski & Peter Ruzicka, editors: SOFSEM 2001: Theory and Practice of Informatics, 28 th Conference on Current Trends in Theory and Practice of Informatics Piestany, Slovak Republic, November 24 - December 1, 2001, Proceedings, Lecture Notes in Computer Science 2234. Springer, pp. 90– 109. Available at: http: //link. springer. de/link/service/series/0558/bibs/22340090. htm. – Jan van Leeuwen & Jirı Wiedermann (2001): The Turing machine paradigm in contemporary computing. In: B. Enquist & W. Schmidt, editors: Mathematics Unlimited — 2001 and Beyond. Springer-Verlag, p. 1139 -1155. – Peter Verbaan, Jan van Leeuwen & Jiri Wiedermann (2004): Complexity of Evolving Interactive Systems. In: Juhani Karhumaki, Hermann A. Maurer, Gheorghe Paun & Grzegorz Rozenberg, editors: Theory Is Forever, Essays Dedicated to Arto Salomaa on the Occasion of His 70 th Birthday, Lecture Notes in Computer Science 3113. Springer, pp. 268– 281. Available at: http: //springerlink. metapress. com/openurl. asp? genre=article& issn=03029743& volume=3113& spage=268. – Jiri Wiedermann (2007): Autopoietic automata: Complexity issues in offspring-producing evolving processes. Theor. Comput. Sci. 383(2 -3), pp. 260– 269.

Advanced topics in Cognitive Computation • Formative (30%): – Two summary and two background seminars on chapter(s) from Aleksander, I. & Morton, H. , (2012), ‘Aristotle’s Laptop: the discovery of our informational mind’, World Scientific. • Summative (70%) – (a) A mini (~500) word review ‘Aristotle’s Laptop: the discovery of our informational mind’, World Scientific (the best short review to be submitted for publication in AISBQ magazine). – (b) A more in-depth (~2000 word) researched review of the work as a whole (the best in depth review to be submitted for publication in Cognitive Computation journal) • Both reviews to be submitted by 5 pm, Friday 22 nd March, 2013; at a minimum both reviews should discuss: (a) what is the core thesis of the text and (b) is thesis convincing. The second (extended) review should also: (a) identify and outline the relationship of the text to other key works in the field and (b) offer at least some [critical] analysis of the ideas.

Seminar lists • • C 2/Week 2: Stephen summary; Emily background C 3/Week 3: Asei summary; Monica background C 4/Week 4: Monica summary; Asei background C 5/Week 5: Emily summary; Stephen background READING WEEK • • • C 6/Week 7: Anna summary; Emily background C 7/Week 8: Monica summary; Asei background C 8/Week 9: Emily summary; Monica background C 9/Week 10: Asei Summary; Stephen background C 10/Week 11: Stephen summary; Anna REVIEW

Research project and Dissertation • The research project is a substantial software implementation in the broad area of Cognitive Computing. • The project aims to allow students to put into practice some of the newly acquired cognitive computing techniques on either a technical implementation - for example the use of AI or Neural Networks in a computer game - or an experimental investigation of some cognitive phenomena (e. g. change blindness). • Selection of an appropriate research project is one aspect of the assessment process.

Requirements • Students implement a Cognitive Computing project and dissertation, which should include a suitable literature search of the research area, emphasising the methodological aspects of the work. • Students should select their project during terms one and two, by liaising with potential research supervisors (check out faculty web pages for research interests) and submit a project form (indicating project supervisors) by the end of term two (the form is provided in the course handbook). • This is a formal requirement for progression; if the project form has not been submitted, then the student will not be allowed to progress to dissertation.

Length etc. • During the third term students should make arrangements to meet with their project supervisor on at least four occasions: twice at the start of the project (to introduce the project; confirm key project questions and strategy etc) and twice towards the end of term three (to confirm thesis outline; demonstrate any project work and/or highlight key experimental findings; review answers to research questions; demonstrate an appropriate academic writing style etc). • The course is primarily assessed via a substantial 15 -20, 000 word dissertation together with an appropriate software implementation and/or series of experiments. • Research only projects can be considered in exceptional circumstances. • The dissertation must be completed by the end of the third term (20 rd September 2013).