The assignment description still contained a few references to Blackboard, which I’ve replaced with those for PolyLearn. In particular, I’ve changed the part for obtaining the environment.

[…]
The emphasis in this assignment lies on the search methods, and not on the programming of the environment. You will use the Bot Environment, a slightly revised version of the environment for this assignment that was also used in previous lab exercises. Alternatively, you can also use the WumpusEnvironment, which will be used again in the second assignment. Both of them are available from PolyLearn under “2 - Intelligent Agents.”
[…]

Tags: Assignment

I have received a request to review an article for the ACM Computing Reviews. It's of moderate interest for me, but if anybody from this class is interested, let me know. This could be the basis for your AI nugget presentation, or use it as extra credit, or for replacing a lab, (partial) assignment or quiz.

F.K.


===================================================================
Review of an already published article for ACM Computing Reviews and ComputingReviews.com (http://www.computingreviews.com)
(This review will be published on the ComputingReviews.com Web site, and is not about acceptance or rejection. It is intended to provide feedback on already published work. )

"Chaotic exploration and learning of locomotion behaviors" Shim Y., Husbands P.
Neural Computation 24(8): 2185-2222,2012

CCS Classification:

Abstract:
We present a general and fully dynamic neural system, which exploits intrinsic chaotic dynamics, for the real-time goal-directed exploration and learning of the possible locomotion patterns of an articulated robot of an arbitrary morphology in an unknown environment. The controller is modeled as a network of neural oscillators that are initially coupled only through physical embodiment, and goal-directed exploration of coordinated motor patterns is achieved by chaotic search using adaptive bifurcation. The phase space of the indirectly coupled neural-body-environment system contains multiple transient or permanent self-organized dynamics, each of which is a candidate for a locomotion behavior. The adaptive bifurcation enables the system orbit to wander through various phase-coordinated states, using its intrinsic chaotic dynamics as a driving force, and stabilizes on to one of the states matching the given goal criteria. In order to improve the sustainability of useful transient patterns, sensory homeostasis has been introduced, which results in an increased diversity of motor outputs, thus achieving multiscale exploration. A rhythmic pattern discovered by this process is memorized and sustained by changing the wiring between initially disconnected oscillators using an adaptive synchronization method. Our results show that the novel neurorobotic system is able to create and learn multiple locomotion behaviors for a wide range of body configurations and physical environments and can readapt in realtime after sustaining damage.
===================================================
Article Review

The article has already been published, and the review is intended for additional feedback; if accepted, the review will appear in the ACM Computing Reviews (deadline is probably around Oct. 18).

My expectations for a review are:
- you read the article (which I will send you); usually I read such articles at least twice
- you will check the Internet for related articles (the ones mentioned in the bibliography section, and possibly others the authors may have omitted); typically I won't read all of those, but I'll look at the abstracts, skim the most relevant ones, and read those that help me understand the one under review
- use your best judgement to evaluate the quality of the article, based on a review form provided by the journal editors
- write a review for people interested in the topic; this one is usually short (about 250 words)

Overall, such a review typically takes me 5-10 hours; count on spending maybe a little more if you're not familiar with the topic. In both cases, I will act as the main reviewer, and reserve the right to use or omit your contributions. I will acknowledge your assistance to the editors. If I use any of your contributions for the ACM Computing Reviews, your name will also be forwarded as a co-author of the review.  In that case, you need to sign a copyright release form.

Tags: Review, Lab, Assignment, AI Nugget

I've changed Assignment 3 to be optional. This means that you can submit A1 and A2 only, and each of them counts for 50% of the assignment grade, or you can submit A1, A2, and A3 for 33% each. I'll calculate the maximum of the two, so by submitting A3, you have nothing to lose.
If you’re considering doing A3, let me know so I have an idea of how many students may take part in the competitions.

Tags: Assignment, Deadline