UNDER CONSTRUCTION
Poole and Mackworth textbook: https://artint.info/
https://www.heardspace.org
http://faculty.virginia.edu/baygame/
- Read Mark O. Riedl and Vadim Bulitko (2013). Interactive Narrative:
An Intelligent Systems Approach, AI Magazine, Spring Issue, 67—77. https://www.aaai.org/ojs/index.php/aimagazine/article/view/2449 - Axelrod, Robert. 1984. The Evolution of Cooperation. New York: Basic Books. Adaptation retrieved from http://www-ee.stanford.edu/~hellman/Breakthrough/book/pdfs/axelrod.pdf
SELECTED READINGS: CS 4269 – Spring 2019 (Kunda)
Professional topics
· Professional ethics
o ACM code of ethics https://www.acm.org/code-of-ethics
o Duhigg article on Google teams – Teamwork and psychological safety https://www.nytimes.com/2016/02/28/magazine/what-google-learned-from-its-quest-to-build-the-perfect-team.html
o Leveson article on Therac 25 – Computer software disaster http://sunnyday.mit.edu/papers/therac.pdf
o “The perks are great” article – Finding an ethical company to work for https://www.wired.com/2016/05/the-perks-are-great-just-dont-ask-us-what-we-do/
o Ted Chiang – Silicon Valley Is Turning Into Its Own Worst Fear https://www.buzzfeednews.com/article/tedchiang/the-real-danger-to-civilization-isnt-ai-its-runaway
o Marketplace – The Price of Profits https://www.marketplace.org/topics/price-profits
o Why Companies Are Becoming B Corporations https://hbr.org/2016/06/why-companies-are-becoming-b-corporations
· Human subjects research
o Leetaru 2016 – article on research ethics in the era of ML and big data https://www.forbes.com/sites/kalevleetaru/2016/06/17/are-research-ethics-obsolete-in-the-era-of-big-data/#ba019aa7aa3d
o Kramer et al 2014 – Facebook emotion manipulation study https://www.pnas.org/content/111/24/8788 (and “editorial expression of concern” from the journal PNAS, a “top” journal, very unusual) https://www.pnas.org/content/111/29/10779.1
· Writing and communication
o Kunda – Guide to Writing with Figures
General AI and machine learning
· AI = representations + search
o Newell and Simon Turing award lecture (in article form) http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.104.2482&rep=rep1&type=pdf
· Machine learning all topics
o Online lectures by Isbell and Littman (Udacity course) https://www.udacity.com/course/machine-learning–ud262
Supervised learning – General methods
· Basic methodology
o Understanding feature engineering – Supervised learning pipeline, and working with features https://towardsdatascience.com/understanding-feature-engineering-part-1-continuous-numeric-data-da4e47099a7b
o Train/test split and cross validation https://towardsdatascience.com/train-test-split-and-cross-validation-in-python-80b61beca4b6
o Confusion matrices and accuracy metrics https://ibug.doc.ic.ac.uk/media/uploads/documents/ml-lecture3-2014.pdf
o Provost and Fawcett 1997 – Paper on ROC curves and limitations https://www.aaai.org/Papers/KDD/1997/KDD97-007.pdf
· Data leakage (contaminating test data with training data)
o Data leakage in ML – high level overview of the problem https://machinelearningmastery.com/data-leakage-machine-learning/
o Kaufman et al 2011 – technical discussion of leakage, what it is, and how to avoid it https://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.365.7769&rep=rep1&type=pdf
o ICML 2013 whale challenge – famous example of data leakage in a Kaggle contest https://www.kaggle.com/c/the-icml-2013-whale-challenge-right-whale-redux/discussion/4865#25839
o Story of a bad train/test split – another interesting (and subtle) example https://engineering.taboola.com/story-of-bad-train-test-split/
· Features
o Text data – pre-processing and feature extraction https://towardsdatascience.com/understanding-feature-engineering-part-3-traditional-methods-for-text-data-f6f7d70acd41
Decision trees
· Overview
o Mitchell 1997 chapter on decision trees, including pruning
o ID3 algorithm from Isbell and Littman lecture notes
o Random forests https://towardsdatascience.com/the-random-forest-algorithm-d457d499ffcd
· Case study on decision tress and autism diagnosis
o Wall et al – two papers that did not follow good ML practices https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0043855 https://www.nature.com/articles/tp201210
o Bone et al – analysis of the above Wall et al papers, including recommendations for doing solid applied ML work https://link.springer.com/article/10.1007/s10803-014-2268-6
Neural networks
· Overview
o Nielsen deep learning textbook – Perceptrons, multilayer networks, backpropagation, and deep networks http://neuralnetworksanddeeplearning.com/
o Krishevsky et al 2012 – the famous (infamous?) ImageNet paper (first big deep learning success) https://papers.nips.cc/paper/4824-imagenet-classification-with-deep-convolutional-neural-networks.pdf
Reinforcement learning
· Overview
o Silver et al 2016 – famous paper on AlphaGo, first AI program to “beat” humans at the game of Go https://www.nature.com/articles/nature16961
Unsupervised learning
· Overview
o Jain 2010 – Data clustering – 50 years beyond K-means https://www.sciencedirect.com/science/article/pii/S0167865509002323