Sunday, July 23, 2017

Weekly Review 23 July 2013

Below are some of the interesting links I Tweeted about recently.

  1. AI will help you do your job better, not replace you, according to Microsoft: https://www.technologyreview.com/s/608272/microsoft-thinks-ai-will-fill-your-blind-spots-not-take-over-your-job/
  2. Using machine learning to map groups of neurons to behaviours in D. melanagaster: https://www.theregister.co.uk/2017/07/14/ai_deciphers_fruit_fly_brain/
  3. How AI is going to increase social inequality: https://www.theverge.com/2017/7/13/15963710/robots-ai-inequality-social-mobility-study
  4. 150+ tutorials on Python and machine learning: https://unsupervisedmethods.com/over-150-of-the-best-machine-learning-nlp-and-python-tutorials-ive-found-ffce2939bd78
  5. The many ways UK industry is using AI: https://www.theregister.co.uk/2017/07/13/british_ai_smbs/
  6. The maths you need to know to master machine learning: http://dataconomy.com/2017/02/mathematics-machine-learning/
  7. A crash-course to machine learning with the Julia language: http://www.kdnuggets.com/2017/07/guerrilla-guide-machine-learning-julia.html
  8. Introduction to using Python for big data: http://dataconomy.com/2016/10/big-data-python/
  9. 3 educational pathways to get into machine learning: https://www.datanami.com/2017/07/10/machine-learning-education-3-paths-get-started/
  10. A brief description of Creative Adversarial Networks: http://www.kdnuggets.com/2017/07/creative-adversarial-network.html
  11. Paper on Creative Adversarial Networks: https://arxiv.org/abs/1706.07068
  12. Online AI soccer players: https://www.cnet.com/news/after-chess-people-are-now-building-ai-to-beat-us-at-soccer/
  13. It's not that the AI are biased, or even that the developers are biased, it's biased data sets that cause problems: https://www.axios.com/algorithms-discriminate-and-big-tech-doesnt-care-2458083425.html
  14. Cargo Cult Computer Science, or, what I think is wrong with AI research: https://computational-intelligence.blogspot.com/2017/07/cargo-cult-computer-science.html
  15. It looks like for the meantime AI will be assisting radiologists rather than replacing them: http://money.cnn.com/2017/07/14/technology/business/radiology-doctors-artificial-intelligence/index.html
  16. When Generative Adversarial Networks start making up their own language: https://www.fastcodesign.com/90132632/ai-is-inventing-its-own-perfect-languages-should-we-let-it
  17. Personally, I find people misusing AI far more frightening than AI itself: https://www.recode.net/2017/7/15/15976744/elon-musk-artificial-intelligence-regulations-ai
  18. Profile of neural network based speech recognition pioneer Tony Robinson: http://www.theregister.co.uk/2017/07/17/tony_robinson_speechmatics/ 
  19. A hierarchy of malicious algorithms: https://www.theguardian.com/technology/2017/jul/16/how-can-we-stop-algorithms-telling-lies 
  20. Why it is cheaper to buy your own hardware for machine learning than use the cloud: http://www.theregister.co.uk/2017/07/17/hardware_for_machine_learning/ 
  21. Determining if a business is ready to utilise AI - 3 questions to ask: https://www.datanami.com/2017/07/14/business-ready-ai-ask-3-questions/
  22. The limits of deep learning: https://venturebeat.com/2017/04/02/understanding-the-limits-of-deep-learning/ 
  23. Speeding-up convolutional neural networks on a Raspberry Pi: http://cv-tricks.com/artificial-intelligence/deep-learning/accelerating-convolutional-neural-networks-on-raspberry-pi/ 
  24. The top 10 programming languages for 2017: http://spectrum.ieee.org/computing/software/the-2017-top-programming-languages 
  25. AI industry hype-it takes more than a bunch of 'if' statements to qualify as an AI: http://www.theregister.co.uk/2017/07/18/it_vendors_mass_over_hype_artificial_intelligence_abilities/ 
  26. A tutorial on Keras TensorFlow: http://cv-tricks.com/tensorflow-tutorial/keras/
  27. Yandex open sources its machine learning library CatBoost: http://www.theregister.co.uk/2017/07/18/yandex_open_sources_machine_learning_library_catboost/ 
  28. Why AI needs to look to neuroscience more: https://www.theverge.com/2017/7/19/1599
  29. I think implants will be interfaces to external tech for a while yet-organ replacements will be bio-engineered: https://www.theverge.com/2017/7/21/15999544/biohacking-finger-magnet-human-augmentation-loss 
  30. Top ten programming languages for employment: http://spectrum.ieee.org/static/interactive-the-top-programming-languages-2017
  31. A lot of businesses call their products AI, but that doesn't make them so: https://www.axios.com/report-many-firms-are-ai-washing-claims-of-intelligent-products-2461919261.html
  32. Using neural networks to control physical rehabilitation equipment: http://spectrum.ieee.org/the-human-os/biomedical/devices/ai-can-help-patients-recover-ability-to-stand-and-walk
  33. Intel's $79 USB 3 deep learning stick: https://www.siliconrepublic.com/machines/intel-movidius-neural-compute-stick
  34. Yet another reason why you shouldn't review papers for Elsevier journals: https://forestsandco.wordpress.com/2017/07/18/thanks-for-your-invitation-to-review-but/

Thursday, July 13, 2017

Cargo Cult Computer Science

I recently attended a presentation by a post-graduate student that I thought was a little bit funny. The presentation was about the experiments they had done on classifying classical music. At the end of the presentation, they proudly declared that algorithm X could identify the composer of a piece (one of Vivaldi, Bach or Mozart), from half a second of music.

The first query I raised was, how many notes are you going to get in half a second? Classical music tends to have a relatively relaxed pace (at least, compared to the music I enjoy) so I doubt there would be more than one or two notes in each sample. The response was, algorithm X is really good at classification so half a second is enough.

The second query I raised was as follows: there was only one piece from each composer in each sample, and the Vivaldi was entirely strings, the Mozart was entirely piano, and the Bach was a mixture of instruments. How do they know that the algorithm didn't just learn to classify instruments?

This is similar to the famous example from the early days of neural networks, when perceptrons were being trained to distinguish photographs that contained images of tanks and those that did not. After some very good results at the start of the project, a second batch of images utterly failed. The reason for that failure was traced to the fact that the photographs with tanks had been developed using a slightly different process to that used to develop the photographs without tanks. That resulted in a slight difference in the overall brightness of the photographs. The neural network had simply learned to distinguish between lighter and darker photographs.

Now, the people who were looking for tanks did one thing right: they tested their algorithm with more data. The post-graduate student at the start of this story didn't do that. They just looked at the results they got, which fit their expectations, and stopped there. That meant that the conclusions they were drawing were not supported by the evidence.

The American physicist Richard Feynman famously spoke of "Cargo Cult Science". This is research that has the superficial form of science, but does not follow the rigor expected of the scientific method.

The scientific method is a process that has developed over many centuries, and requires a certain rigor and self-criticism that is intended to prevent erroneous conclusions being made. It requires scientists to be completely honest with themselves, to consider every objection to their research method and possible factors that could be influencing their results. The scientific method is supposed to prevent researchers from just seeing what they want to see and instead see the reality. The post-graduate student did not do this, and so their conclusions are not necessarily valid.

I've seen this in a lot of papers in computer science, and in more than a few post-graduate theses. Experiments are performed, results are gathered, and conclusions are confidently espoused about the value of their approach. Yet they never consider what else could explain those results. They never consider whether their data is biased in some way, or if their method is flawed so that certain results are favoured over others.

I think there are several reasons this occurs. Students in computer science are not necessarily trained in the scientific method, so they can hardly be blamed for not following it. It is human nature that researchers want their approach, their new algorithm, to work, so they develop a kind of wilful blindness to the flaws in their experimental approach. Finally, and more insidiously, researchers are under immense pressure to publish: "publish or perish" applies in computer science just as much as any other field of academia. It is only through publishing papers that researchers gain employment, get promotion, and secure research funding. There is, then, a system set up to favour rapid and uncritical publication of supportive results and to suppress unfavourable results. We have created a system that favours Cargo Cult Computer Science.

Computer science, if it is to remain worthy of the appellation "science" must fully embrace the scientific method. This means being rigorous, and being self-critical. The consequences of not doing so, could be severe for everyone in the field.

Weekly Review 13 July 2017

Below are some of the interesting links I Tweeted about recently.

  1. When AI research collides head-on with privacy laws: https://www.theregister.co.uk/2017/07/03/google_deepmind_trial_failed_to_comply_with_data_protection_law/
  2. A list of logical fallacies used in arguments - make sure you don't fall for these in your research: https://www.lifehacker.com.au/2017/06/spot-the-flaw-in-a-politicians-argument-with-this-guide-to-logical-fallacies/
  3. Chinese tech sector's concerns about AI: https://www.technologyreview.com/s/608183/chinas-tech-moguls-warn-of-ais-troubling-trajectory/
  4. Should the heads of universities be academics? https://www.insidehighered.com/news/2017/07/03/madison-professors-fight-keep-requirement-administrators-be-academics
  5. Using AI to assist with drug design: https://www.theregister.co.uk/2017/07/03/gsk_signs_ai_deal_british_firm/
  6. Ten simple rules to make your data science /AI research reproducible: http://dataconomy.com/2017/07/10-rules-results-data-science/ 
  7. Relax, machine learning is not going to revolutionise business just yet: http://www.theregister.co.uk/2017/07/05/rethink_machine_learning/ 
  8. I am starting to think that ethics in AI is a more complicated problem than AI itself: http://www.theregister.co.uk/2017/07/04/ai_ethics_and_what_next/ 
  9. Using deep neural networks to observe football games and bet on them: https://www.theverge.com/2017/7/6/15923784/ai-predict-sport-betting-gambling-stratagem 
  10. In two years every successful software product will be expected to include some form of AI: http://www.techrepublic.com/article/why-there-will-be-no-more-ai-startups-in-2-years/ 
  11. Automating machine learning so non-technical people can use it: https://techcrunch.com/2017/07/06/h2o-ais-driverless-ai-automates-machine-learning-for-businesses/ 
  12. Using machine learning to detect hot jupiter planets orbiting other stars: http://www.theregister.co.uk/2017/07/07/machine_learning_algos_and_hot_jupiters/ 
  13. Google is funding the development of machine-written news articles: https://www.theverge.com/2017/7/7/15933224/google-press-association-ai-news-writers 
  14. Google's unified deep learning model: https://www.datanami.com/2017/07/06/google-mimics-human-brain-unified-deep-learning-model/
  15. A session with an AI therapist: http://www.theregister.co.uk/2017/07/07/ai_robodoc_asks_me_personal_questions/ 
  16. Why machine learning doesn't work for some problems: http://www.datasciencecentral.com/profiles/blogs/the-e-dimension-why-machine-learning-doesn-t-work-well-for-some 
  17. Preparing for the future of ubiquitous machine learning: https://www.datanami.com/2017/07/03/machine-learning-everywhere-preparing-future/ 
  18. List of articles on TensorFlow: http://www.datasciencecentral.com/profiles/blogs/9-great-articles-about-tensorflow 
  19. Another attempt at using a deep neural network to name colours: https://arstechnica.com/information-technology/2017/07/new-experiments-reveal-that-ai-are-still-terrible-at-naming-paint-colors/ 
  20. Getting into deep learning: how they built the Not Hotdog app: http://www.theregister.co.uk/2017/07/10/skills_for_ai/ 
  21. How machine learning / AI is being used in US hospitals: https://www.techemergence.com/top-5-hospitals-using-machine-learning/
  22.  How AI can help to manage data: http://www.techrepublic.com/article/how-ai-and-machine-learning-can-help-solve-its-data-management-problem/
  23. Why AI will change the business world: http://informationweek.com/big-data/software-platforms/artificial-intelligence-will-redefine-the-world/a/d-id/1329254
  24. List of resources on self-driving cars: http://www.kdnuggets.com/2017/07/5-free-resources-getting-started-self-driving-vehicles.html
  25. The ten IT jobs most likely to be taken over by AI: http://informationweek.com/strategic-cio/10-it-jobs-that-will-be-done-by-ai/d/d-id/1329310?
  26. Detecting heart arrhythmia using deep learning: https://www.technologyreview.com/s/608234/the-machines-are-getting-ready-to-play-doctor/
  27. AI is now, finally, starting to deliver in the enterprise: http://computerworld.com/article/3203029/artificial-intelligence/a-i-starts-to-deliver-in-the-enterprise-at-last.html
  28. DeepMind is letting virtual figures teach themselves how to walk, run etc through a virtual environment: https://www.theverge.com/tldr/2017/7/10/15946542/deepmind-parkour-agent-reinforcement-learning
  29. How AI is being used in mobile apps: http://computerworld.com/article/3204906/mobile-wireless/here-s-what-really-matters-when-it-comes-to-ai-and-mobile-apps.html
  30. Detecting heart arrhythmia using deep learning: https://www.technologyreview.com/s/608234/the-machines-are-getting-ready-to-play-doctor/ 
  31. Google is looking seriously at bias in AI: https://www.theverge.com/2017/7/10/15947358/google-ai-pair-bias-fairness-equality
  32. The role of AI in the 'digital brain' of a business: http://www.kdnuggets.com/2017/07/why-every-company-needs-digital-brain.html 
  33. AI is moving further into the enterprise: https://www.datanami.com/2017/07/11/ai-makes-inroads-enterprise-software/
     

Sunday, July 2, 2017

IEEE Transactions on Neural Networks and Learning Systems, Volume 28, Issue 7, July 2017.

1. Feature Selection Based on Structured Sparsity: A Comprehensive Study
Author(s): Jie Gui; Zhenan Sun; Shuiwang Ji; Dacheng Tao; Tieniu Tan
Pages: 1490 - 1507

2. Modeling Disease Progression via Multisource Multitask Learners: A Case Study With Alzheimer’s Disease
Author(s): Liqiang Nie; Luming Zhang; Lei Meng; Xuemeng Song; Xiaojun Chang; Xuelong Li
Pages: 1508 - 1519

3. Observer-Based Adaptive NN Control for a Class of Uncertain Nonlinear Systems With Nonsymmetric Input Saturation
Author(s): Yong-Feng Gao; Xi-Ming Sun; Changyun Wen; Wei Wang
Pages: 1520 - 1530

4. Neural Approximation-Based Adaptive Control for a Class of Nonlinear Nonstrict Feedback Discrete-Time Systems
Author(s): Yan-Jun Liu; Shu Li; Shaocheng Tong; C. L. Philip Chen
Pages: 1531 - 1541

5. Characterization of Linearly Separable Boolean Functions: A Graph-Theoretic Perspective
Author(s): Yanyi Rao; Xianda Zhang
Pages: 1542 - 1549

6. Fine-Grained Image Classification via Low-Rank Sparse Coding With General and Class-Specific Codebooks
Author(s): Chunjie Zhang; Chao Liang; Liang Li; Jing Liu; Qingming Huang; Qi Tian
Pages: 1550 - 1559

7. Impulsive Multisynchronization of Coupled Multistable Neural Networks With Time-Varying Delay
Author(s): Yan-Wu Wang; Wu Yang; Jiang-Wen Xiao; Zhi-Gang Zeng
Pages: 1560 - 1571

8. Embedded Streaming Deep Neural Networks Accelerator With Applications
Author(s): Aysegul Dundar; Jonghoon Jin; Berin Martini; Eugenio Culurciello
Pages: 1572 - 1583

9. Solution Path for Pin-SVM Classifiers With Positive and Negative τ Values
Author(s): Xiaolin Huang; Lei Shi; Johan A. K. Suykens
Pages: 1584 - 1593

10. Adaptive Event-Triggered Control Based on Heuristic Dynamic Programming for Nonlinear Discrete-Time Systems
Author(s): Lu Dong; Xiangnan Zhong; Changyin Sun; Haibo He
Pages: 1594 - 1605

11. An Alternating Identification Algorithm for a Class of Nonlinear Dynamical Systems
Author(s): Yajun Zhang; Tianyou Chai; Dianhui Wang
Pages: 1606 - 1617

12. Exponential Synchronization for Markovian Stochastic Coupled Neural Networks of Neutral-Type via Adaptive Feedback Control
Author(s): Huabin Chen; Peng Shi; Cheng-Chew Lim
Pages: 1618 - 1632

13. Observer-Based Adaptive Neural Network Trajectory Tracking Control for Remotely Operated Vehicle
Author(s): Zhenzhong Chu; Daqi Zhu; Simon X. Yang
Pages: 1633 - 1645

14. Structural Minimax Probability Machine
Author(s): Bin Gu; Xingming Sun; Victor S. Sheng
Pages: 1646 - 1656

15. Global Synchronization of Multiple Recurrent Neural Networks With Time Delays via Impulsive Interactions
Author(s): Shaofu Yang; Zhenyuan Guo; Jun Wang
Pages: 1657 - 1667

16. Batch Mode Active Learning for Regression With Expected Model Change
Author(s): Wenbin Cai; Muhan Zhang; Ya Zhang
Pages: 1668 - 1681

17. Prediction Reweighting for Domain Adaptation
Author(s): Shuang Li; Shiji Song; Gao Huang
Pages: 1682 - 1695

18. Switched-Observer-Based Adaptive Neural Control of MIMO Switched Nonlinear Systems With Unknown Control Gains
Author(s): Lijun Long; Jun Zhao
Pages: 1696 - 1709

19. μ-Stability of Nonlinear Positive Systems With Unbounded Time-Varying Delays
Author(s): Tianping Chen; Xiwei Liu
Pages: 1710 - 1715

20. Learning With Auxiliary Less-Noisy Labels
Author(s): Yunyan Duan; Ou Wu
Pages: 1716 - 1721

21. Import Vector Domain Description: A Kernel Logistic One-Class Learning Algorithm
Author(s): Sergio Decherchi; Walter Rocchia
Pages: 1722 - 1729

Saturday, July 1, 2017

Weekly Review 1 July 2017

Below are some of the interesting links I Tweeted about recently.
  1. What Garry Kasparov thinks of AI: http://www.bbc.com/future/story/20170616-garry-kasparov-why-the-world-should-embrace-ai
  2. Top 15 Python libraries for data science: http://www.kdnuggets.com/2017/06/top-15-python-libraries-data-science.html
  3. A lot of money is getting invested in AI: https://siliconangle.com/blog/2017/06/17/ai-explodes-investors-pour-big-bucks-startups/
  4. A collection of cheat sheets for machine learning and deep learning libraries: https://medium.com/@kailashahirwar/essential-cheat-sheets-for-machine-learning-and-deep-learning-researchers-efb6a8ebd2e5
  5. The General AI Challenge has a $5M prize: http://spectrum.ieee.org/tech-talk/robotics/artificial-intelligence/in-the-general-ai-challenge-teams-compete-for-5-million
  6. A method of measuring progress in AI research: https://www.eff.org/ai/metrics
  7. What it means for humans to be smart, in the age of AI: https://hbr.org/2017/06/in-the-ai-age-being-smart-will-mean-something-completely-different
  8. My father was in his mid-40s when I was born: https://www.theguardian.com/science/2017/jun/20/older-men-fathers-geekier-sons-study-geek-index
  9. I've seen a few cover letters with pretty bad errors in them: http://www.nzherald.co.nz/business/news/article.cfm?c_id=3&objectid=11879982
  10. It seems like the only people opposed to sci-hub are the journal publishers: https://www.nature.com/news/us-court-grants-elsevier-millions-in-damages-from-sci-hub-1.22196
  11. TEU is being disingenuous - NZ universities have been profit-driven for decades: http://www.nzherald.co.nz/nz/news/article.cfm?c_id=1&objectid=11884020
  12. A brief overview of machine learning: http://www.kdnuggets.com/2017/06/making-sense-machine-learning.html
  13. Using machine learning to clean data before applying machine learning to it: http://dataconomy.com/2017/06/machine-learning-for-enterprise/
  14. List of five deep learning code demos: http://www.kdnuggets.com/2017/06/deep-learning-demos-code-beginners.html
  15. Neural networks explained using a corporate analogy: https://myabakhova.blogspot.com/2017/06/
  16. Why you should never answer questions about salary in a job interview: http://www.nzherald.co.nz/business/news/article.cfm?c_id=3&objectid=11883755
  17. The killer business apps of deep learning: http://informationweek.com/big-data/what-execs-should-know-about-deep-learning/a/d-id/1329196?
  18. A series of tutorials on the TensorFlow API: http://www.kdnuggets.com/2017/06/using-tensorflow-api-tutorial-series.html
  19. Using AI to get business insights: http://dataconomy.com/2017/06/business-insights-artificial-intelligence/
  20. Potential applications of AI in banking: http://informationweek.com/big-data/how-artificial-intelligence-will-revolutionize-banking/a/d-id/1329218?
  21. How ratty data messes up machine learning algorithms: http://dataconomy.com/2017/06/faulty-data-machine-learning/
  22. Finding how deep neural networks arrive at decisions: https://techcrunch.com/2017/06/30/mit-csail-research-offers-a-fully-automated-way-to-peer-inside-neural-nets/
  23. Enabling sales with machine learning: https://www.techemergence.com/machine-learning-in-sales-enablement-applications/
  24. Brief example of web scraping with R: http://www.kdnuggets.com/2017/06/web-scraping-r-online-food-blogs.html
  25. Instagram is using machine learning to filter hurtful comments: https://techcrunch.com/2017/06/29/instagram-implements-an-ai-system-to-fight-mean-and-harassing-comments/
  26. Trust your head, not your gut, your gut is the one part of your body that's guaranteed to be full of cr*p: http://www.kdnuggets.com/2017/06/who-cares-evidence.html
  27. Might to time to start holding these kinds of competitions in countries other than the USA: https://www.theverge.com/2017/6/30/15903206/afghanistan-robot-team-denied-us-visa
  28. The WCCI 2018 call for papers is out: http://www.ecomp.poli.br/~wcci2018/
  29. Generative adversarial networks producing new styles of art: https://www.newscientist.com/article/2139184-artificially-intelligent-painters-invent-new-styles-of-art/
  30. Richard Feynman's essay on cargo cult science: http://calteches.library.caltech.edu/51/2/CargoCult.htm

   

Thursday, June 29, 2017

IEEE Transactions on Fuzzy Systems, Volume 25, Issue 3, June 2017

1. A Profit Maximizing Solid Transportation Model Under a Rough Interval Approach
Author(s): Amrit- Das, Uttam Kumar Bera, and Manoranjan Maiti
Pages: 485-498

2. Design of State Feedback Adaptive Fuzzy Controllers for Second-Order Systems Using a Frequency Stability Criterion
Author: Krzysztof Wiktorowic
Pages: 499-510

3. Dynamic Output-Feedback Dissipative Control for T–S Fuzzy Systems With Time-Varying InputDelay and Output Constraints
Author(s): Hyun Duck Choi, Choon Ki Ahn, Peng Shi, Ligang Wu and Myo Taeg Lim
Pages: 511-526

4. Adaptive Predefined Performance Control for MIMO Systems With Unknown Direction via Generalized Fuzzy Hyperbolic Model
Author(s): Lei Liu, Zhanshan Wang, Zhanjun Huang, and Huaguang Zhang
Pages: 527-542

5. Revisiting Fuzzy Set and Fuzzy Arithmetic Operators and Constructing New Operators in the Land of Probabilistic Linguistic Computing
Author: Shing-Chung Ngan
Pages: 543-555

6. Asymptotic Fuzzy Tracking Control for a Class of Stochastic Strict-Feedback Systems
Author(s): Ci Chen, Zhi Liu, Yun Zhang, C. L. Philip Chen,and Shengli Xie
Pages: 556-568

7. Approaches to T–S Fuzzy-Affine-Model-Based Reliable Output Feedback Control for Nonlinear Itˆo Stochastic Systems
Author(s): Yanling Wei, Jianbin Qiu, Hak-Keung Lam, and Ligang Wu
Pages: 569-583

8. Pixel Modeling Using Histograms Based on Fuzzy Partitions for Dynamic Background Subtraction
Author(s): Zhi Zeng, Jianyuan Jia, Dalin Yu, Yilong Chen, and Zhaofei Zhu
Pages: 584-593

9. Varying Spread Fuzzy Regression for Affective Quality Estimation
Author(s): Kit Yan Chan and Ulrich Engelke
Pages: 594-613

10. Ranking of Multidimensional Uncertain Information Based on Metrics on the Fuzzy EllipsoidNumber Space
Author(s): Guixiang Wang and Yun Li
Pages: 614-626

11. The Spatial Disaggregation Problem: Simulating Reasoning Using a Fuzzy Inference System
Author: J¨org Verstraete
Pages: 627-641

12. Adaptive Fuzzy Backstepping Tracking Control for Strict-Feedback Systems With Input Delay
Author(s): Hongyi Li, Lijie Wang, Haiping Du, and Abdesselem Boulkroune
Pages: 642-652

13. Dynamic Output Feedback-Predictive Control of a Takagi–Sugeno Model With Bounded Disturbance
Author(s): Baocang Ding and Hongguang Pan
Pages: 653-667

14. Command-Filtered-Based Fuzzy Adaptive Control Design for MIMO-Switched Nonstrict-Feedback Nonlinear Systems
Author(s): Yongming Li and Shaocheng Tong
Pages: 668-681

15. Type-2 Fuzzy Alpha-Cuts
Author(s): Hussam Hamrawi, Simon Coupland, and Robert John
Pages: 682-692

16. A New Look at Type-2 Fuzzy Sets and Type-2 Fuzzy Logic Systems
Author: Li-Xin Wang
Pages: 693-706

17. LMI-Based Stability Analysis for Piecewise Multi-affine Systems
Author(s): Anh-Tu Nguyen, Michio Sugeno, Victor Campos, and Michel Dambrine
Pages: 707-714

18. An Extended Type-Reduction Method for General Type-2 Fuzzy Sets
Author(s): Bing-Kun Xie and Shie-Jue Lee
Pages: 715-724

19. Critique of "A New Look at Type-2 Fuzzy Sets and Type-2 Fuzzy Logic Systems"
Author(s): Jerry M. Mendel and Dongrui Wu
Pages: 725-727

Monday, June 19, 2017

Weekly Review 19 June 2017

Below are some of the interesting links I Tweeted about recently.

  1. An Ask Slashdot discussion on jobs in AI: https://ask.slashdot.org/story/17/06/09/1954230/ask-slashdot-what-types-of-jobs-are-opening-up-in-the-new-field-of-ai
  2. A script-writing AI producing lines for David Hasselhoff: https://arstechnica.com/the-multiverse/2017/04/an-ai-wrote-all-of-david-hasselhoffs-lines-in-this-demented-short-film/
  3. Creating a classifier in SKLearn: http://www.datasciencecentral.com/profiles/blogs/creating-your-first-machine-learning-classifier-model-in-sklearn 
  4. Five things AI (neural networks) can do better than people: https://www.datanami.com/2017/06/08/5-things-ai-better/
  5. Using machine learning to detect identity thieves, from the way they move their mouse: https://qz.com/1003221/identity-theft-can-be-thwarted-by-artificial-intelligence-analysis-of-a-users-mouse-movements/
  6. Combining quantum computing and machine learning: http://spectrum.ieee.org/tech-talk/robotics/artificial-intelligence/kickstarting-the-quantum-startup-a-hybrid-of-quantum-computing-and-machine-learning-is-spawning-new-ventures
  7. Predicting suicide risk with machine learning: https://qz.com/1001968/artificial-intelligence-can-now-predict-suicide-with-remarkable-accuracy/
  8. A neural network based negotiating bot: https://qz.com/1004070/facebook-fb-built-an-ai-system-that-learned-to-lie-to-get-what-it-wants/
  9. Automating comment moderation with machine learning: https://www.recode.net/2017/6/13/15789178/new-york-times-expanding-comments-artificial-intelligence-google
  10. A deep learning powered four armed robot that composes marimba music: http://spectrum.ieee.org/automaton/robotics/artificial-intelligence/four-armed-marimba-robot-uses-deep-learning-to-compose-its-own-music
  11. Yes, prospective employers will check your social media presences: https://www.seek.co.nz/career-advice/recruiters-reveal-top-3-social-media-fails
  12. A collection of 150 intelligent agents has beaten the game Ms Pacman: https://techcrunch.com/2017/06/15/microsofts-ai-beats-ms-pac-man/
  13. 8 techniques for data mining: https://www.datanami.com/2017/06/14/8-concrete-data-mining-techniques-will-deliver-best-results/
  14. Microsoft is using deep neural networks to turn photos into art: https://techcrunch.com/2017/06/15/microsoft-pix-can-now-turn-your-iphone-photos-into-art-thanks-to-a-i/
  15. Applying AI to call centres: http://www.techproresearch.com/article/how-artificial-intelligence-is-taking-call-centers-to-the-next-level/
  16. AI is predicted to create 800,000 jobs by 2021: http://www.techrepublic.com/article/can-ai-really-create-800000-jobs-by-2021-this-report-says-yes/

Tuesday, June 13, 2017

IEEE Transactions on Cognitive and Developmental Systems, Volume 9, Number 2, June 2017

1. Yielding Self-Perception in Robots Through Sensorimotor Contingencies
Author(s): P. Lanillos, E. Dean-Leon and G. Cheng
Pages: 100 - 112

2. Online Multimodal Ensemble Learning Using Self-Learned Sensorimotor Representations
Author(s): M. Zambelli and Y. Demirisy
Pages: 113 - 126

3. Perception of Localized Features During Robotic Sensorimotor Development
Author(s): A. Giagkos, D. Lewkowicz, P. Shaw, S. Kumar, M. Lee and Q. Shen
Pages: 113 - 126

4. Building a Sensorimotor Representation of a Naive Agent’s Tactile Space
Author(s): V. Marcel, S. Argentieri and B. Gas
Pages: 141 - 152

5. A Multimodal Model of Object Deformation Under Robotic Pushing
Author(s): V. E. Arriola-Rios and J. L. Wyatt
Pages: 153 - 169

6. Analysis of Cognitive Dissonance and Overload through Ability-Demand Gap Models
Author(s): G. Hossain and M. Yeasin
Pages: 170 - 182

7. Constructing a Language From Scratch: Combining Bottom–Up and Top–Down Learning Processes in a Computational Model of Language Acquisition
Author(s): J. Gaspers, P. Cimiano, K. Rohlfing and B. Wrede
Pages: 183 - 196

8. Behavior-Based SSVEP Hierarchical Architecture for Telepresence Control of Humanoid Robot to Achieve Full-Body Movement
Author(s): J. Zhao, W. Li, X. Mao, H. Hu, L. Niu and G. Chen
Pages: 197 - 209


Monday, June 12, 2017

IEEE Transactions on Fuzzy System, Volume 25, Issue 3, June 2017

1. A Profit Maximizing Solid Transportation Model Under a Rough Interval Approach
Author(s): A. Das, U. Kumar Bera and M. Maiti
Pages: 485-498

2. Design of State Feedback Adaptive Fuzzy Controllers for Second-Order Systems Using a Frequency Stability Criterion
Author(s): K. Wiktorowicz
Pages: 499-510

3. Dynamic Output-Feedback Dissipative Control for T–S Fuzzy Systems With Time-Varying Input Delay and Output Constraints
Author(s): H. D. Choi, C. K. Ahn, P. Shi, L. Wu and M. T. Lim
Pages: 511-526

4. Adaptive Predefined Performance Control for MIMO Systems With Unknown Direction via Generalized Fuzzy Hyperbolic Model
Author(s): L. Liu, Z. Wang, Z. Huang and H. Zhang
Pages: 527-542

5. Revisiting Fuzzy Set and Fuzzy Arithmetic Operators and Constructing New Operators in the Land of Probabilistic Linguistic Computing
Author(s): S. C. Ngan
Pages: 543-555

6. Asymptotic Fuzzy Tracking Control for a Class of Stochastic Strict-Feedback Systems
Author(s): C. Chen, Z. Liu, Y. Zhang, C. L. P. Chen and S. Xie
Pages: 556-568

7. Approaches to T–S Fuzzy-Affine-Model-Based Reliable Output Feedback Control for Nonlinear Itô Stochastic Systems
Author(s): Y. Wei, J. Qiu, H. K. Lam and L. Wu
Pages: 569-583

8. Pixel Modeling Using Histograms Based on Fuzzy Partitions for Dynamic Background Subtraction
Author(s): Z. Zeng, J. Jia, D. Yu, Y. Chen and Z. Zhu
Pages: 584-593

9. Varying Spread Fuzzy Regression for Affective Quality Estimation
Author(s): K. Y. Chan and U. Engelke
Pages: 594-613

10. Ranking of Multidimensional Uncertain Information Based on Metrics on the Fuzzy Ellipsoid Number Space
Author(s): G. Wang and Y. Li
Pages: 614-626

11. The Spatial Disaggregation Problem: Simulating Reasoning Using a Fuzzy Inference System
Author(s): J. Verstraete
Pages: 627-641

12. Adaptive Fuzzy Backstepping Tracking Control for Strict-Feedback Systems With Input Delay
Author(s): H. Li, L. Wang, H. Du and A. Boulkroune
Pages: 642-652

13. Dynamic Output Feedback-Predictive Control of a Takagi–Sugeno Model With Bounded Disturbance
Author(s): B. Ding and H. Pan
Pages: 653-667

14. Command-Filtered-Based Fuzzy Adaptive Control Design for MIMO-Switched Nonstrict-Feedback Nonlinear Systems
Author(s): Y. Li and S. Tong
Pages: 668-681

15. Type-2 Fuzzy Alpha-Cuts
Author(s): H. Hamrawi, S. Coupland and R. John
Pages: 682-692

16. A New Look at Type-2 Fuzzy Sets and Type-2 Fuzzy Logic Systems
Author(s): L. X. Wang
Pages: 693-706

17. LMI-Based Stability Analysis for Piecewise Multi-affine Systems
Author(s): A. T. Nguyen, M. Sugeno, V. Campos and M. Dambrine
Pages: 707-714

18. An Extended Type-Reduction Method for General Type-2 Fuzzy Sets
Author(s): B. K. Xie and S. J. Lee
Pages: 715-724

19. Critique of “A New Look at Type-2 Fuzzy Sets and Type-2 Fuzzy Logic Systems"
Author(s): J. M. Mendel and D. Wu
Pages: 725-727

Saturday, June 10, 2017

Weekly Review 9 June 2017

Below are some of the interesting links I Tweeted about recently.

  1.  Microsoft has released version 2.0 of its deep learning toolkit: https://techcrunch.com/2017/06/01/microsoft-releases-version-2-0-of-its-deep-learning-toolkit/
  2. An overview of tensors, as used in TensorFlow: http://www.kdnuggets.com/2017/06/deep-learning-demystifying-tensors.html 
  3. How rat brains are inspiring the next generation of artificial neural networks: http://spectrum.ieee.org/biomedical/imaging/ai-designers-find-inspiration-in-rat-brains
  4. Despite the article title Watson isn't diagnosing cancer, it's helping plan how to treat it: https://news.fastcompany.com/ibm-says-watson-healths-ai-is-getting-really-good-at-diagnosing-cancer-4039394
  5. An overview of neurocomputing hardware - hardware inspired by the brain: http://spectrum.ieee.org/computing/hardware/the-brain-as-computer-bad-at-math-good-at-everything-else
  6. Four tips for job interviews - these apply to interviews for academic positions as well: https://www.seek.co.nz/career-advice/4-ways-to-prepare-for-an-interview
  7. Three things intelligent machines need to copy from the human brain: http://spectrum.ieee.org/computing/software/what-intelligent-machines-need-to-learn-from-the-neocortex
  8. Hawkins' book "On Intelligence" is very good, must re-read sometime soon: http://spectrum.ieee.org/computing/software/what-intelligent-machines-need-to-learn-from-the-neocortex
  9. So valedictorians (dux in NZ) go on to lead good lives, but seldom achieve particularly highly: http://time.com/money/4779223/valedictorian-success-research-barking-up-wrong/
  10. Why we need to build an artificial brain or, why neuromorphic computing is a good idea: http://spectrum.ieee.org/computing/hardware/can-we-copy-the-brain
  11. We could build an artificial brain now, but probably wouldn't want to pay for it: http://spectrum.ieee.org/computing/hardware/we-could-build-an-artificial-brain-right-now
  12. If you're looking for a job, be careful what you say on social media (including Twitter!) http://jobs.ieee.org/jobs/content/Looking-for-a-Job-Double-Check-Your-Social-Media-Accounts-First-2017-04-18
  13. Deep learning can't tell what Homer Simpson is doing. Or, why DeepMind built an enormous database of videos: http://spectrum.ieee.org/tech-talk/robotics/artificial-intelligence/deepmind-shows-ai-has-trouble-seeing-homer-simpson-actions
  14. If I am advertising for a new staff member, I will read the cover letters of applicants: https://www.seek.co.nz/career-advice/does-anyone-actually-read-cover-letters
  15. Using AI in children's educational apps: https://techcrunch.com/2017/06/07/sesame-workshop-and-ibm-team-up-to-test-a-new-a-i-powered-teaching-method/
  16. Combining relational networks with deep learning: https://www.theregister.co.uk/2017/06/09/deepmind_teaches_ai_to_reason/
  17. Paper on adding relational reasoning to deep neural networks: https://arxiv.org/abs/1706.01427
  18. How AI can make things worse, instead of better: http://www.kdnuggets.com/2017/06/unintended-consequences-machine-learning.html

Friday, June 9, 2017

Neural Networks, Volume 92, Pages 1-98, August 2017

Special Issue "Advances in Cognitive Engineering Using Neural Networks" Edited by Minho Lee, Steven Bressler and Robert Kozma

1. Advances in Cognitive Engineering Using Neural Networks  
Author(s): Minho Lee, Steven Bressler, Robert Kozma
Pages: 1-2

2. How can a recurrent neurodynamic predictive coding model cope with fluctuation in temporal patterns? Robotic experiments on imitative interaction   
Author(s): Ahmadreza Ahmadi, Jun Tani
Pages: 3-16

3. Dual-memory neural networks for modeling cognitive activities of humans via wearable sensors   
Author(s): Sang-Woo Lee, Chung-Yeon Lee, Dong-Hyun Kwak, Jung-Woo Ha, Jeonghee Kim, Byoung-Tak Zhang
Pages: 17-28

4. Understanding human intention by connecting perception and action learning in artificial agents   
Author(s): Sangwook Kim, Zhibin Yu, Minho Lee
Pages: 29-38

5. Application of structured support vector machine backpropagation to a convolutional neural network for human pose estimation   
Author(s): Peerajak Witoonchart, Prabhas Chongstitvatana
Pages: 39-46

6. Reprint of “Two-stage sparse coding of region covariance via Log-Euclidean kernels to detect saliency”   
Author(s): Ying-Ying Zhang, Cai Yang, Ping Zhang
Pages: 47-59

7. Evaluating deep learning architectures for Speech Emotion Recognition   
Author(s): Haytham M. Fayek, Margaret Lech, Lawrence Cavedon
Pages: 60-68

8. Weighted spatial based geometric scheme as an efficient algorithm for analyzing single-trial EEGS to improve cue-based BCI classification   
Author(s): Fatemeh Alimardani, Reza Boostani, Benjamin Blankertz
Pages: 69-76

9. Prediction of advertisement preference by fusing EEG response and sentiment analysis   
Author(s): Himaanshu Gauba, Pradeep Kumar, Partha Pratim Roy, Priyanka Singh, Debi Prosad Dogra, Balasubramanian Raman
Pages: 77-88

10. A Hypergraph and Arithmetic Residue-based Probabilistic Neural Network for classification in Intrusion Detection Systems   
Author(s): M.R. Gauthama Raman, Nivethitha Somu, Kannan Kirthivasan, V.S. Shankar Sriram
Pages: 89-97