Computational Intelligence

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

1. 7 trends in AI for 2016: http://www.techrepublic.com/article/7-trends-for-artificial-intelligence-in-2016-like-2015-on-steroids/
2. Learn deep learning from Google using TensorFlow http://googleresearch.blogspot.co.nz/2016/01/teach-yourself-deep-learning-with.html
3. Learning to code neural networks http://www.kdnuggets.com/2016/01/learning-to-code-neural-networks.html
4. The parts for building an AI assistant are available online, mostly for free, according to this article: http://www.techrepublic.com/article/ai-helpers-arent-just-for-facebooks-zuckerberg-heres-how-to-build-your-own/
5. Microsoft open sources it's deep learning toolkit: http://blogs.microsoft.com/next/2016/01/25/microsoft-releases-cntk-its-open-source-deep-learning-toolkit-on-github/
6. Damn, Marvin Minsky died: http://www.nytimes.com/2016/01/26/business/marvin-minsky-pioneer-in-artificial-intelligence-dies-at-88.html?_r=0
7. Deep Feelings on Deep Learning: http://www.kdnuggets.com/2016/01/deep-feelings-deep-learning.html … it all seems a bit like magic
8. 7 mistakes in data science, apply to computational intelligence and data mining too: http://www.kdnuggets.com/2016/01/7-common-data-science-mistakes.html
9. Video courses in deep learning and machine learning: http://www.datasciencecentral.com/profiles/blogs/step-by-step-video-courses-for-deep-learning-and-machine-learning
10. Implementing k nearest neighbours in Python: http://www.kdnuggets.com/2016/01/implementing-your-own-knn-using-python.html
11. Random thoughts on machine learning and AI: http://www.datasciencecentral.com/profiles/blogs/big-data-analytics-data-science-machine-learning-random-insights
12. Overview of recurrent neural networks: http://spectrum.ieee.org/computing/software/the-neural-network-that-remembers
13. "Impostor Syndrome" among academics: http://www.nature.com/naturejobs/science/articles/10.1038/nj7587-555a?WT.mc_id=TWT_NatureNews
14. Deep learning with Spark and TensorFlow: http://www.kdnuggets.com/2016/01/deep-learning-spark-tensorflow.html
15. 12 things not to do when applying for academic jobs https://www.insidehighered.com/advice/2016/01/29/common-mistakes-academic-job-seekers-make-essay … I'd add don't go in to an interview with a bad attitude
16. Doesn't matter how good you are, if you have a bad or arrogant attitude in the interview, you won't get the job https://www.insidehighered.com/advice/2016/01/29/common-mistakes-academic-job-seekers-make-essay
17. Deep learning neural network with human-level performance playing Go http://spectrum.ieee.org/tech-talk/computing/software/monster-machine-defeats-prominent-pro-player?utm_campaign=Weekly%20Notification-%20IEEE%20Spectrum%20Tech%20Alert&utm_source=boomtrain&utm_medium=email&utm_term=555a972628fbca1d260da1ba&utm_content=Monster%20Machine%20Cracks%20The%20Game%20Of%20Go&bt_alias=eyJ1c2VySWQiOiIzMGJmOTkxMy1mOThiLTQ5YTgtOTIzMy1iYmMzODk4ZDcxODcifQ%3D%3D

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

1. Step-by-step tutorial for MS Azure ML environment: http://www.kdnuggets.com/2016/01/guide-azure-machine-learning-studio.html
2. ANN + GA foe FOREX trading, via @gcosma1 http://forexpost.org/determinants-of-exchange-rates/neural-network-genetic-algorithm-in-forex-trading/ … Fine as a tutorial, can think of several ways of doing this better...
3. How not to write about science: https://theconversation.com/how-not-to-write-about-science-52202
4. The future of AI: http://www.datasciencecentral.com/profiles/blogs/the-future-of-artificial-intelligence-is-here - bad news for low-skilled workers
5. Tutorials on the Python machine learning toolkit scikit-learn http://www.kdnuggets.com/2016/01/scikit-learn-tutorials-introduction-classifiers.html
6. Brief overview of fuzzy matching algorithms http://www.datasciencecentral.com/profiles/blogs/fuzzy-matching-algorithms-to-help-data-scientists-match-similar
7. The unreasonable reputation of neural networks: http://thinkingmachines.mit.edu/blog/unreasonable-reputation-neural-networks
8. AI will soon be putting middle-skilled workers out of a job - http://www.techrepublic.com/article/more-money-for-the-rich-fewer-jobs-for-everyone-else-the-price-of-the-coming-ai-revolution/?tag=nl.e101&s_cid=e101&ttag=e101&ftag=TRE684d531&utm_content=buffer8e45f&utm_medium=social&utm_source=twitter.com&utm_campaign=buffer - bad news for clerks, sales, and support staff

Been away on my summer vacation the last few weeks, so not been able to blog much. Some interesting links that I Tweeted about since the last review a month ago:

1. Using machine learning in an NFL confidence pool http://theinstitute.ieee.org/ieee-roundup/opinions/ieee-roundup/how-to-use-machine-learning-to-beat-your-friends-in-an-nfl-confidence-pool
2. How eBay enterprise uses machine learning to detect fraudsters http://www.datanami.com/2015/12/21/tis-the-season-to-hunt-fraudsters-with-big-data/
3. Free data mining software:http://www.datasciencecentral.com/profiles/blogs/4-packages-for-data-analysis - R and Weka are there of course, hadn't heard of Orange before.
4. List of some real-world machine learning data sets: http://www.kdnuggets.com/2015/12/tour-real-world-machine-learning-problems.html The academic ones are all really well known (that is, old)
5. datasciencecentral.com/profiles/blogs/internet-of-things-selected-articles The Internet of Things is become really important in computational intelligence - so much data to model!
6. 15 words you shouldn't use if you want to sound smarter-I'd be happy if people stopped confusing "infer" and "imply" http://mashable.com/2015/05/03/words-eliminate-vocabulary/?utm_cid=p-disp-fb%23lHxBPJzBTRqW#kMQHWVmxvGqw
7. Big Data in agriculture - CI has a big role to play in agro/ecol data processing as well: http://www.datasciencecentral.com/profiles/blogs/big-data-in-agriculture-ddw2-1
8. Getting your paper noticed: http://blogs.nature.com/naturejobs/2016/01/06/five-top-tips-for-getting-your-paper-noticed … One of the five points is using social media: http://www.fasttrackimpact.com/#!Create-a-social-media-strategy-for-your-research-that-delivers-real-impact/hmlp3/564df9090cf20af044b924ca
9. Bias is a potential problem in all data sets, not just Big Data: http://www.datanami.com/2016/01/08/beware-of-bias-in-big-data-feds-warn/
10. Applying ANN to proteomics: https://agenda.weforum.org/2015/12/how-machine-learning-helps-biologists-crack-lifes-secrets/ Something I was looking at about 15 years ago...
11. 5 papers on Deep Learning explained: http://www.kdnuggets.com/2016/01/more-arxiv-deep-learning-papers-explained.html
12. Yahoo releases a 1.5 TB data set: https://thestack.com/cloud/2016/01/14/yahoo-news-dataset-artificial-intelligence-news-feed/
13. The differences between machine learning, machine intelligence, deep learning and AI: http://www.kdnuggets.com/2016/01/what-is-machine-intelligence-ml-deep-learning-ai.html
14. Finding whales in ocean photographs, a step-by-step tutorial: http://www.datasciencecentral.com/profiles/blogs/finding-whales-in-ocean-water-edge-detection-blob-processing-and
15. Deep learning projects on GitHub: http://www.kdnuggets.com/2016/01/top-10-deep-learning-github.html
16. Having kids is a disadvantage in a research career: http://www.sciencedaily.com/releases/2016/01/160111092607.htm - I'd rather have my daughter than a high-powered career
17. AI is set to wipe out a lot of casual and low-skilled jobs: http://www.spectator.co.uk/2016/01/i-robot-you-unemployed/
18. Machine Intelligence in the Real World-how companies go to market: http://techcrunch.com/2015/11/26/machine-intelligence-in-the-real-world/
19. Uploading a paper to http://academia.org gives more citations over time: https://www.academia.edu/12297791/Open_Access_Meets_Discoverability_Citations_to_Articles_Posted_to_Academia.edu Can't cite a paper that can't be found

Some interesting links that I Tweeted about this week:

1. 10 Deep Learning Tips & Tricks http://www.kdnuggets.com/2015/12/top-10-deep-learning-tips-tricks.html These apply to most learning systems, especially cross-validation of data.
2. Is coding ability mostly down to natural aptitude? http://mikehadlow.blogspot.co.nz/2015/12/learn-to-code-its-harder-than-you-think.html I've always found it easy, but have seen so many others struggle
3. Top 10 machine learning projects on GitHub http://www.kdnuggets.com/2015/12/top-10-machine-learning-github.html
4. Deep learning and topological data analysis https://www.linkedin.com/pulse/6-crazy-things-deep-learning-topological-data-can-do-your-kibardin
5. Improvements to Google's TensorFlow open source machine learning package: http://www.kdnuggets.com/2015/12/update-google-tensorflow-deep-learning-is-improving.html
6. 33 Big Data predictions for 2016: http://www.datanami.com/2015/12/15/industry-speaks-top-33-big-data-predictions-for-2016/
7. Updated list of 50 deep learning tools: http://www.kdnuggets.com/2015/12/deep-learning-tools.html

Some interesting links that I Tweeted about this week:

• 5 open-source ML systems from big companies: http://www.datasciencecentral.com/profiles/blogs/5-machine-learning-open-source-projects-from-top-internet
• Facebook open sources its AI hardware design: https://code.facebook.com/posts/1687861518126048/facebook-to-open-source-ai-hardware-design/ … not sure how AI this is, however.
• Cloud-based training of deep-learning neural networks: http://www.datanami.com/2015/12/07/10027/
• List of 20 big data repositories http://www.datasciencecentral.com/profiles/blogs/20-big-data-repositories-you-should-check-out-1
• 20 lessons from building machine learning systems http://www.kdnuggets.com/2015/12/xamat-20-lessons-building-machine-learning-systems.html
• Doctor Andy's Rule Language (DARL) for creating fuzzy systems http://www.docandys.com/ How does this compare to FML? https://en.wikipedia.org/wiki/Fuzzy_markup_language
• Top machine learning algorithms over time http://www.datasciencecentral.com/profiles/blogs/top-10-machine-learning-algorithms
• Detecting how wet a road is using a microphone and deep learning http://arxiv.org/pdf/1511.07035v2.pdf
• List of 50 machine learning API http://www.kdnuggets.com/2015/12/machine-learning-data-science-apis.html

Some interesting links that I Tweeted about this week:

• Piece in the NZ Herald about NeuLab at KEDRI, http://www.nzherald.co.nz/technology/news/article.cfm?c_id=5&objectid=11554230
• Alleged exam cheat goes to extremes to cover up her deeds at Otago Uni: http://www.nzherald.co.nz/nz/news/article.cfm?c_id=1&objectid=11553961
• Using deep neural networks for art forgery: http://www.kdnuggets.com/2015/12/deep-learning-art-forgery.html
• Ten free machine learning books. Most have PDF versions to download. http://www.datasciencecentral.com/profiles/blogs/10-free-machine-learning-books
• Interactive visualisation of basic backpropagation trained MLP http://mwskirpan.com/NN_viz/
• 4 data processing and mining tools from Google http://www.datasciencecentral.com/profiles/blogs/4-open-source-cloud-machine-learning-data-analytics-visualization
• Machine learning in JavaScript http://www.datasciencecentral.com/profiles/blogs/machine-learning-in-javascript-a-compilation-of-resources … Should I include JavaScript output in my SECoS compiler? http://ecos.watts.net.nz/Software/SECoSTools.html
• 11 applications of machine learning: http://www.informationweek.com/strategic-cio/executive-insights-and-innovation/11-cool-ways-to-use-machine-learning/d/d-id/1323375

Some interesting links that I Tweeted about this week:

• My review on Evolving Connectionist Systems has been cited 60 times now https://scholar.google.com/citations?view_op=view_citation&hl=en&user=Z29KBKYAAAAJ&citation_for_view=Z29KBKYAAAAJ:9yKSN-GCB0IC Two more years and I'll write an update
• Another comparison of Python and R for data science and analytics: http://www.datasciencecentral.com/profiles/blogs/programming-python-vs-r
• CV and interview pitfalls for academic job applicants: https://www.insidehighered.com/advice/2015/11/23/easy-fixes-your-cv-r%C3%A9sum%C3%A9-and-interview-answers-essay
• Predicting the gender of Tweeters using machine learning: http://www.kdnuggets.com/2015/11/machine-learning-predict-gender.html
• Science, and reality, is neither left- or right-wing, it just is. Politicians need to understand that: https://www.newscientist.com/article/dn28556-new-twist-in-republican-war-on-climate-science-is-unbelievable/
• Detecting fraudulent in-app purchases with decision trees: http://www.kdnuggets.com/2015/11/detecting-app-purchase-fraud-machine-learning.html What happens to the user if the decision tree is wrong?
• On the increasing prominence of machine learning - seems to have been enabled by the cloud https://channels.theinnovationenterprise.com/articles/are-we-approaching-the-golden-age-of-machine-learning

I've not been posting to this blog very much recently. This is mostly because I have been putting more of my energy into posting interesting information on Twitter. My Twitter profile is here:

https://twitter.com/DrMikeWatts


I tweet there almost every day. Below is a list of my computational intelligence-related Tweets from the last week.

• 7 steps to machine learning in Python: http://www.kdnuggets.com/2015/11/seven-steps-machine-learning-python.html
• Social media for academics: https://medium.com/advice-and-help-in-authoring-a-phd-or-non-fiction/are-you-an-academic-hermit-6d7ae5a0f16a
• Predicting student success - http://www.datasciencecentral.com/profiles/blogs/predictive-analytics-goes-to-college-to-predict-student-success
• Waste disposal, lawsuit selection and parking enforcement, 3 real-world applications of machine learning: http://www.datanami.com/2015/11/18/three-unique-ways-machine-learning-is-being-used-in-the-real-world/
• List of deep-learning tools: http://www.kdnuggets.com/2015/06/popular-deep-learning-tools.html
• Another cloud-based deep-learning system: http://www.datanami.com/2015/11/17/cloud-based-deep-learning-tool-explains-predictions/
• Predicting online gambling addiction using machine learning https://thestack.com/cloud/2015/10/26/machine-learning-to-help-predict-online-gambling-addiction/
• The role of machine learning in cybersecurity http://www.datanami.com/2015/11/16/machine-learnings-big-role-in-the-future-of-cybersecurity/ … I know of several papers applying my SECoS algorithm to cybersecurity
• Experiments with AWS machine learning: http://www.datasciencecentral.com/profiles/blogs/experimenting-with-aws-machine-learning-for-classification
• Microsoft open sources part of their machine learning technology: https://thestack.com/world/2015/11/13/microsoft-open-sources-its-machine-learning-toolkit/
• Neural storytellers: http://www.kdnuggets.com/2015/11/samim-recurrent-neural-net-describe-images-taylor-swift.html
• List of JavaScript machine learning libraries: http://www.datasciencecentral.com/profiles/blogs/machine-learning-in-javascript-a-compilation-of-resources
• Many middle-class jobs are now being threatened by AI. How long before higher education goes the same way? http://www.theguardian.com/technology/2015/nov/12/thinking-machines-skilled-job-robots-steal When an AI can teach a class of undergrads, I'll be worried
• 7 free machine learning courses: http://www.datasciencecentral.com/profiles/blogs/7-free-machine-learning-courses
• Natural Language Processing with Convolutional Neural Networks: http://www.kdnuggets.com/2015/11/understanding-convolutional-neural-networks-nlp.html
• Facebook is also investing big in machine learning / artificial intelligence: http://www.businessinsider.com.au/facebook-outlines-its-artificial-and-machine-learning-ambitions-2015-11
• TensorFlow, a neural network / machine intelligence toolkit from Google, now open source: https://tensorflow.org
• A report on some first experiences with TensorFlow: http://www.kdnuggets.com/2015/11/google-tensorflow-deep-learning-disappoints.html The author seems to be under-whelmed

Atom

Atom