Back to my home page; to my teaching page.

Biologically Inspired Computing F21BC2: 2015/16

General Notes on Course Delivery 

Note:    The lecture slides and other materials that are scheduled later than today’s date are not necessarily complete and/or correct.

DC (David Corne) Lectures are green PV (Patricia Vargas) lectures are blue ML (Michael Lones) lectures are yellow

A White lecture slot means that no class is planned – however sometimes these slots may be used, for example to replace lectures that had to be cancelled. In those cases, we will of course let you know as far as possible in advance.

DC Office hour Thursdays 4:15—5:15

PV Office hour: Tuesdays 2:15---3:15 (To be confirmed)

ML Office hour: Wednesday 10:15—11:15

The SCHEDULE below is correct, but the ‘FUTURE’ Lecture and Coursework material is always subject to change

Week beginning	Tuesday 16:15 in EM 3.06	Thursday 15:15 in EM 3.06	Coursework out	Coursework in	Mandatory additional material	Recommended reading
Mon 14^th Sep	DC Overview of module and intro to Bio-Inspired Computing - ppt	DC Evolutionary Computation ppt	DC hands out coursework 1 (10% of module) CW1 (ppt)		Hard and Easy optimization problems Study the upcoming encodings lecture to help with CW1	simple interactive evolution simple EA/TSP demo (not quite finished)
Mon 21^st Sep	PV Introduction to Neurocomputation (pdf)	PV Artificial Neural Networks I (pdf)				Neural computation tutorial questions and bibliography
Mon 28^th Sep	DC EAs: Details, Encodings, Operators and CW1 ppt	PV Artificial Neural Networks II (pdf)		coursework 1.1 Deadline 23:59pm Sun 4^th Oct	more encodings ppt Selection ppt Operators ppt	Freitas paper pdf Falkenauer paper pdf
Mon 5th Oct	PV Artificial Neural Networks IV (pdf)	DC Hillclimbing, Landscapes, Neighbourhoods, Local Search ppt			Operators ( as above) ppt – describing line/box crossover & Gaussian mut	Here is the SNNS ANN simulator http://www.ra.cs.uni-tuebingen.de/SNNS/ and some further info from Patricia Selected EA Applications paper pdf
Mon 12th Oct		DC Indirect encodings and hyper-heuristics ppt			Reynolds’ paper	A nice explanation of Reynolds’ rules DC’s PSO app  (currently broken) Swarm Intelligence chapter pdf
Mon 19th Oct	PV Artificial Neural Networks V (pdf)	PV GasNets lecture (pdf) Renan Moioli’s invited GasNets lecture pdf	PV hands out coursework 2 (20% of module) CW2 pdf CW2 Data (zip)	coursework 1.2 Deadline 23:59pm Sun 25^th Oct
Mon 26^th Oct	DC Swarm Intelligence ppt	DC Ant Colony Optimization ppt
Mon 2^nd Nov	ML Foundations of Genetic Programming pdf	ML GP: Memory and loops pdf	ML hands out coursework 3 (ON VISION) (20% of module) cw3 pdf zip of ecj
Mon 9th Nov	ML GP: Language and Representation pdf	ML: Cellular automata pdf		Handin coursework 2 Deadline: HANDIN to School Office by 3pm 16^th November		DC’s CA / forest fire app DC’s 1dca app CA urbanization paper CA traffic simulation paper CA flu paper CA brain tumour paper CA and P Systems for HIV
Mon 16th Nov	ML Gene regulatory models pdf	ML: Evolvable hardware (possibly guest lecture)
Mon 23rd Nov	ML Artifical biochemical networks pdf	DC Particle Swarm Optimization ppt
Mon 30^th Nov	Slot to be used if necessary	DC revision lecture	)	Handin Coursework 3 Deadline 3pm Friday 4^th December at School Office

DC = David Corne, who will generally lecture about bio-inspired methods for optimisation, with a focus on evolutionary computation (aka genetic algorithms) – broadly this is about how certain aspects of nature (evolution, swarm behaviour) lead to very effective optimisation and design methods.

PV = Patricia Vargas, who will generally lecture about neural computation, and some popular and effective combinations of neural and evolutionary computation, including evolutionary robotics.

Coursework Handin : CW1 handin in all cases should be by email with the coursework as a ppt or pdf attachment, emailled to DWC at dwcorne@gmail.com, subject line either: “BIC CW1 Slide X” . Include your name in the pdf!

Feel completely free to email any of us at any time about any query regarding the lecture material – you will appreciate that you may rarely get an immediate reply, but we will try to get back to you as soon as possible, and arrange meetings to discuss issues where necessary.

Assessment

50% coursework and 50% exam.

CW 1 (DC): 10% CW2 (pv): 20% CW3 (ML): 20%

As in previous years, the exam will have 4 questions, and you are required to answer only 3 questions for full marks.

Here are some Past papers: 2012, 2013, 2014 . Note that: we used to teach ‘DNA computing’ on this module, but we don’t teach that now. Obviously, you will therefore not get questions like Q1(b) or Q4 in the 2012 paper . Also, this year is the first time that Michael Lones has been involved; so, your paper may or may not involve questions based on the ML lectures.

Here are some notes about the exam, and about revising for DWC’s part of the course: pdf