Weekly Review 20 May 2016

Some interesting links that I Tweeted about in the last week:

1. Three skills data scientists need: http://www.kdnuggets.com/2016/05/practical-skills-practical-data-scientists-need.html
2. A pathway to malevolent AI: http://www.techrepublic.com/article/creating-malevolent-ai-a-manual/
3. What to do if ANN error increases: http://www.kdnuggets.com/2016/05/troubleshooting-neural-network-error-increase.html
4. Deep learning ANN in self-driving cars: http://www.informationweek.com/mobile/mobile-devices/nvidia-car-learns-to-drive-by-watching-humans/d/d-id/1325460?
5. Reproducing van Gogh's painting style with deep learning neural networks: https://www.technologyreview.com/s/601424/algorithm-clones-van-goghs-artistic-style-and-pastes-it-onto-other-images-movies/
6. Machine learning system that gives the conditions for growing new types of crystals: http://futurism.com/machine-learning-uses-human-failures-to-make-crystals/
7. Do computers in classrooms lower exam performance? http://www.theregister.co.uk/2016/05/12/mit_study_finds_students_assisted_by_computers_do_worse_in_exams/
8. Speeding-up neural networks by doing fewer multiplications: http://arxiv.org/abs/1510.03009
9. On de-coupling peer review from specific journals: https://www.insidehighered.com/views/2016/05/16/why-not-make-academic-journal-acceptance-portable-essay
10. More about machine learning in materials science: http://nextbigfuture.com/2016/05/machine-learning-techniques-could.html
11. Perhaps open review would reduce the tendency that anonymous reviewers have to be dicks: https://www.insidehighered.com/views/2016/05/16/open-peer-review-journal-articles-offers-significant-benefits-essay
12. How long before these AI are writing student essays-for-hire? http://www.theverge.com/2016/5/15/11678142/google-ai-writes-fiction-natural-language-neural-network
13. First materials science, now an AI does physics: http://www.eurekalert.org/pub_releases/2016-05/anu-air051316.php?utm_source=dlvr.it&utm_medium=twitter
14. This article seems to be confusing Elm the programming language with ELM as in Extreme Learning Machines: http://www.valuewalk.com/2016/04/future-machine-learning/
15. An introduction to natural language processing, with some useful links to information and libraries: http://blog.algorithmia.com/2016/04/introduction-to-natural-language-processing/
16. Badder than a bad thing that's very, very bad: http://motherboard.vice.com/en_au/read/elsevier-buys-ssrn
17. I've certainly encountered my share of narcissists in academia: http://www.theguardian.com/education/2016/may/17/university-research-academic-bragging-grants
18. The case for randomly accepting borderline papers: http://www.kdnuggets.com/2016/05/embrace-random-acceptance-borderline-papers.html
19. How and why machine learning isn't enough in financial fraud detection: http://dataconomy.com/machine-learning-fraud-artificial-intelligence-isnt-enough/
20. Semi-supervised reinforcement learning: http://www.kdnuggets.com/2016/05/intro-semi-supervised-reinforcement-learning.html
21. Some resources on deep learning: http://www.datasciencecentral.com/profiles/blogs/deep-learning-definition-resources-comparison-with-machine-learni
22. How can we control an AI if nobody understands it? http://techcrunch.com/2016/05/16/how-can-we-control-intelligent-systems-no-one-fully-understands/
23. Twitter has developed an AI that can recognise what is happening in videos: https://www.technologyreview.com/s/601284/twitters-artificial-intelligence-knows-whats-happening-in-live-video-clips/
24. Seems like GoButler is offering a natural-language processing service for hire: http://techcrunch.com/2016/05/16/angel-ai/
25. Intelligent chatbots for banking customer service: https://www.technologyreview.com/s/601418/do-your-banking-with-a-chatbot/
26. Will machine learning bring about the end of coding? http://www.wired.com/2016/05/the-end-of-code/
27. Google has created its own ASIC chips to implement deep neural networks: http://www.wired.com/2016/05/google-tpu-custom-chips/
28. Claims that Google's deep neural network chip could advance Moore's law by 7 years: http://www.pcworld.com/article/3072256/google-io/googles-tensor-processing-unit-said-to-advance-moores-law-seven-years-into-the-future.html#comments
29. 12 ways AI could disrupt the senior executives of a corporation: http://www.informationweek.com/big-data/12-ways-ai-will-disrupt-your-c-suite/d/d-id/1325557?
30. An overview of word2vec, encoding words to vectors: http://www.kdnuggets.com/2016/05/amazing-power-word-vectors.html
31. List of lists of resources on machine learning, deep learning, and natural language processing: http://www.datasciencecentral.com/profiles/blogs/curated-lists-of-data-science-machine-learning-deep-learning-and
32. A neural-network based approach for finding a photo that most matches a sketch: https://www.newscientist.com/article/mg23030742-600-scan-your-doodles-to-find-the-perfect-matching-photo-online/
33. Yahoo's meme-GIF making AI: http://motherboard.vice.com/en_au/read/these-fire-gifs-were-made-by-artificial-intelligence-yahoo
34. Some niche machine learning software projects: http://www.kdnuggets.com/2016/05/five-machine-learning-projects-cant-overlook.html
35. Google's Awareness API: http://www.theverge.com/2016/5/19/11712608/android-awareness-api-google-io-2016
36. Description of 3 clustering algorithms, k-means, EM clustering and Affinity Propagation: https://www.toptal.com/machine-learning/clustering-algorithms
37. Some supposed progress towards artificial general intelligence: http://nextbigfuture.com/2016/05/vicarious-will-show-off-their-progress.html