AI, explain yourself !

“It’s time for AI to move out its adolescent, game-playing phase and take seriously the notions of quality and reliability.”

There is an interesting commentary with interviews by Don MONROE in the recent Communications of the ACM, November 2018, Volume 61, Number 11, Pages 11-13, doi:

Artificial Intelligence (AI) systems are taking over a vast array of tasks that previously depended on human expertise and judgment (only). Often, however, the “reasoning” behind their actions is unclear, and can produce surprising errors or reinforce biased processes. One way to address this issue is to make AI “explainable” to humans—for example, designers who can improve it or let users better know when to trust it. Although the best styles of explanation for different purposes are still being studied, they will profoundly shape how future AI is used.

Some explainable AI, or XAI, has long been familiar, as part of online recommender systems: book purchasers or movie viewers see suggestions for additional selections described as having certain similar attributes, or being chosen by similar users. The stakes are low, however, and occasional misfires are easily ignored, with or without these explanations.

“Considering the internal complexity of modern AI, it may seem unreasonable to hope for a human-scale explanation of its decision-making rationale”.

Read the full article here:
https://cacm.acm.org/magazines/2018/11/232193-ai-explain-yourself/fulltext

 

 

What if the AI answers are wrong?

Cartoon no. 1838 from the xkcd [1] Web comic by Randall MUNROE [2] describes in a brilliant sarcastic way the state of the art in AI/machine learning today and shows us the current main problem directly. Of course you will always get results from one of your machine learning models. Just fill in your data and you will get results – any results. That’s easy. The main question remains open: “What if the results are wrong?” The central problem is to know at all that my results are wrong and to what degree. Do you know your error? Or do you just believe what you get? This can be ignored in some areas, desired in other areas, but in a safety critical domain, e.g. in the medical area, this is crucial [3]. Here also the interactive machine learning approach can help to compensate or lower the generalization error through human intuition [4].

 

[1] https://xkcd.com

[2] https://en.wikipedia.org/wiki/Randall_Munroe

[3] Andreas Holzinger, Chris Biemann, Constantinos S. Pattichis & Douglas B. Kell 2017. What do we need to build explainable AI systems for the medical domain? arXiv:1712.09923. online available: https://arxiv.org/abs/1712.09923v1

[4] Andreas Holzinger 2016. Interactive Machine Learning for Health Informatics: When do we need the human-in-the-loop? Brain Informatics, 3, (2), 119-131, doi:10.1007/s40708-016-0042-6. online available, see:
https://hci-kdd.org/2018/01/29/iml-human-loop-mentioned-among-10-coolest-applications-machine-learning

There is also a discussion on the image above:

https://www.explainxkcd.com/wiki/index.php/1838:_Machine_Learning

 

 

Microsoft boosts Explainable AI

Microsoft invests into explainable AI and acquired on June, 20, 2018 Bonsai, a California start-up, which was founded by Mark HAMMOND and Keen BROWNE in 2014. Watch an excellent introduction “Programming your way to explainable AI” by Mark HAMMOND here:

and read read the original story about the acquisition here:

https://blogs.microsoft.com/blog/2018/06/20/microsoft-to-acquire-bonsai-in-move-to-build-brains-for-autonomous-systems

“No one really knows how the most advanced algorithms do what they do. That could be a problem.” Will KNIGHT in “The dark secret of the heart of AI”

https://www.technologyreview.com/s/604087/the-dark-secret-at-the-heart-of-ai

NEW: The Travelling Snakesman v 1.1 – 18.4.2018 released

Enjoy the new version of our travelling snakesman game:
https://hci-kdd.org/gamification-interactive-machine-learning/

Please follow the instructions given. By playing this game you help to proof the following hypothesis:
“A human-in-the-loop enhances the performance of an automatic algorithm”

Human-in-the-loop AI

Human-in-the-Loop-AI

This is really very interesting. In the recent April, 5, 2018, TWiML & AI (This Week in Machine Learning and Artificial Intelligence) podcast, Robert MUNRO (a graduate from Stanford University, who is an recognized expert in combining human and machine intelligence) reports on the newly branded Figure Eight [1] company, formerly known as CrowdFlower. Their Human-in-the-Loop AI platform supports data science & machine learning teams working on various topics, including autonomous vehicles, consumer product identification, natural language processing, search relevance, intelligent chatbots, and more. Most recently on disaster response and epidemiology. This is a further proof on the enormous importance and potential usefulness of the human-in-the-loop interactive machine Leanring (iML) approach! Listen to this awesome discussion led excellently by Sam CHARRINGTON:

https://twimlai.com/twiml-talk-125-human-loop-ai-emergency-response-robert-munro/

This discussion fits well to the previous discussion with Jeff DEAN (head of the Google Brain team) – who emphasized the importance of health and the limits of automatic approaches including deep learning. Enjoy to listen directly at:

https://twimlai.com/twiml-talk-124-systems-software-machine-learning-scale-jeff-dean/

[1] https://www.figure-eight.com/resources/human-in-the-loop

 

A good proof of the importance of the HCI-KDD approach, worth: 2,1 Billion USD !

Our strategic aim is to find solutions for data intensive problems by the combination of two areas, which bring ideal pre-conditions towards understanding intelligence and to bring business value in AI: Human-Computer Interaction (HCI) and Knowledge Discovery (KDD). HCI deals with questions of human intelligence, whereas KDD deals with questions of artificial intelligence, in particular with the development of scalable algorithms for finding previously unknown relationships in data, thus centers on automatic computational methods. A proverb attributed perhaps incorrectly to Albert Einstein illustrates this perfectly: “Computers are incredibly fast, accurate, but stupid. Humans are incredibly slow, inaccurate, but brilliant. Together they may be powerful beyond imagination” [1].

An article published on February, 18, 2018 by David Shaywitz [2] from Forbes reports on the recent purchase of  the oncolology data company Flatiron Health for the enormous sum of 2,1 Billion USD (remember: Deep Mind was purchased by Google for a mere 400 million GBP 😉

This supports a few hypotheses which I try to convince my students all the time (but they won’t believe me unless Google is doing it 😉

a) those who can turn raw health data into insights and understandable knowledge can produce value
b) data – and particularly big data – is useless for the decision maker, what they need is reliable, valuable and trustworthy information
c) for the complexity of sensemaking from health data we (still) need a human-in-the-loop:  Humans (still) exceed machine performance in understanding the context and explaining the underlying explanatory factors of the data
d) consequently this is a good example for the business value of our HCI-KDD approach: Let the computer find in arbitrarily high-dimensional spaces what no human is able to do – but let the human do what no computer is able to do: BOTH working together are powerful beyond imagination!

Flatiron Health [3] is a company which is specialized on health data curation, supported by technology of course, but mostly done manually by human experts in the Mechanical Turk style. Remark: The name mechanical turk has historic origins as it was inspired by an automatic 18th-century chess-playing machine by Wolfgang von Kempelen,  that beats e.g. Benjamin Franklin in chess playing – and was acclaimed as “AI”. However, ti was later revealed that it was neither a machine nor an automatic device – in fact it was a human chess master hidden in a secret space under the chessboard and controlling the movements of an humanoid dummy. Similarly,  services which help to solve problems via human intelligence are called “Mechanical Turk online services”.

[1] Holzinger, A. 2013. Human–Computer Interaction and Knowledge Discovery (HCI-KDD): What is the benefit of bringing those two fields to work together? In: Cuzzocrea, Alfredo, Kittl, Christian, Simos, Dimitris E., Weippl, Edgar & Xu, Lida (eds.) Multidisciplinary Research and Practice for Information Systems, Springer Lecture Notes in Computer Science LNCS 8127. Heidelberg, Berlin, New York: Springer, pp. 319-328, doi:10.1007/978-3-642-40511-2_22

[2] https://www.forbes.com/sites/davidshaywitz/2018/02/18/the-deeply-human-core-of-roches-2-1b-tech-acquisition-and-why-they-did-it/#6242fdbc29c2

[3] https://flatiron.com

On-Device Machine Intelligence

One very interesting approach of federated machine learning is presented by Sujith Ravi from Google: Machine learning models (e.g. CNN) are sucessfully used for the design of intelligent systems capable of visual recognition, speech and language understanding. Most of these are running on a cloud – which is often inpredictable where it is physically running. A huge problem so far is that typical machine learning models are awkward to use on mobile devices due to both computational and memory constraints. While these devices could make use of models running on high-performance data centers with CPUs or GPUs, this is not feasible for many applications and scenarios where inference needs to be performed directly “on” device. This requires re-thinking existing machine learning algorithms and coming up with new models that are directly optimized for on-device machine intelligence rather than doing post-hoc model compression. Sujith Ravi is introducing a novel “projection-based” machine learning system for training compact neural networks. The approach uses a joint optimization framework to simultaneously train a “full” deep network and a lightweight “projection” network. Unlike the full deep network, the projection network uses random projection operations that are efficient to compute and operates in bit space yielding a low memory footprint. The system is trained end-to-end using backpropagation. Ravi shows that the approach is flexible and easily extensible to other machine learning paradigms, for example, they can learn graph-based projection models using label propagation. The trained “projection” models are then directly used for inference, please watch the origial video on:

 

Python in Machine Learning still Nr. 1 and increasing

There is of course no such thing like a ‘best language for machine learning’ – but as a matter of fact Python is still Nr. 1 and increasing:
Image Source: https://stackoverflow.blog/2017/09/06/incredible-growth-python/

We use in all our courses Python due to the fact that it is an “industrial standard” and widely available. I would love e.g. Julia, which is much faster, but it remains rather academic and needs a lot of additional effort. It is not astonishing that Python is worldwide the most popular tool for machine learning and artificial intelligence as there are deep learning frameworks available, including Tensor Flow, Pandas, NumPy, PyBrain, Scikit, SimpleAI, EasyAI, etc. etc.

Consequently, in our courses we teach Python, have a look at:

Marcus D. Bloice & Andreas Holzinger 2016. A Tutorial on Machine Learning and Data Science Tools with Python. In: Holzinger, Andreas (ed.) Machine Learning for Health Informatics, Lecture Notes in Artificial Intelligence LNAI 9605. Heidelberg: Springer, pp. 437-483, doi:10.1007/978-3-319-50478-0_22. [link to paper]

iML with the human-in-the-loop mentioned among 10 coolest applications of machine learning

Within the “Two Minute Papers” series, Karol Károly Zsolnai-Fehér from the Institute of Computer Graphics and Algorithms at the Vienna University of Technology mentions among “10 even cooler Deep Learning Applications” our human-in-the-loop paper:

Seid Muhie Yimam, Chris Biemann, Ljiljana Majnaric, Šefket Šabanović & Andreas Holzinger 2016. An adaptive annotation approach for biomedical entity and relation recognition. Springer/Nature: Brain Informatics, 3, (3), 157-168, doi:10.1007/s40708-016-0036-4

Watch the video here (iML is mentinoned from approx. 1:20):

Here the list of all 10 papers discussed within this 2-minutes-video

1. Geolocation – http://arxiv.org/abs/1602.05314
2. Super-resolution – http://arxiv.org/pdf/1511.04491v1.pdf
3. Neural Network visualizer – http://experiments.mostafa.io/public/…
4. Recurrent neural network for sentence completion:
5. Human-in-the-loop and Doctor-in-the-loop: https://link.springer.com/article/10.1007/s40708-016-0036-4
6. Emoji suggestions for images – https://emojini.curalate.com/
7. MNIST handwritten numbers in HD – http://blog.otoro.net/2016/04/01/generating-large-images-from-latent-vectors
8. Deep Learning solution to the Netflix prize – https://karthkk.wordpress.com/2016/03/22/deep-learning-solution-for-netflix-prize/
9. Curating works of art –
10. More robust neural networks against adversarial examples – http://cs231n.stanford.edu/reports201…
The Keras library: http://keras.io/

A) The basic principle of the iML human-in-the-loop approach:

Andreas Holzinger 2016. Interactive Machine Learning for Health Informatics: When do we need the human-in-the-loop? Brain Informatics, 3, (2), 119-131, doi:10.1007/s40708-016-0042-6

B) The entry in the GI Lexikon:
https://gi.de/informatiklexikon/interactive-machine-learning-iml

C) The experimental proof-of-concept:

Andreas Holzinger, Markus Plass, Katharina Holzinger, Gloria Cerasela Crisan, Camelia-M. Pintea & Vasile Palade 2017. A glass-box interactive machine learning approach for solving NP-hard problems with the human-in-the-loop. arXiv:1708.01104.

D) Outline and Survey of application possibilities:

Andreas Holzinger, Chris Biemann, Constantinos S. Pattichis & Douglas B. Kell 2017. What do we need to build explainable AI systems for the medical domain? arXiv:1712.09923.

Andreas Holzinger, Bernd Malle, Peter Kieseberg, Peter M. Roth, Heimo Müller, Robert Reihs & Kurt Zatloukal 2017. Towards the Augmented Pathologist: Challenges of Explainable-AI in Digital Pathology. arXiv:1712.06657.

 

NIPS-2017 Best paper “Explainability was one of the major reasons the paper was given the award”

Congratulations to Arthur GRETTON from the Gatsby Computational Neuroscience Unit at the University College London an his team. Their paper titled “A Linear-Time Kernel Goodness-of-Fit Test” authored by Wittawat JITKRITTUM, Wenkai XU, Zoltan SZABO, Kenji FUKUMIZU and Arthur GRETTON won the prestigous NIPS 2017 best paper award. In the interview by Sam Charringtion from TWiML&AI, the authors of the NIPS 2017 best paper said at 14:10 in the following video that ” … explainability was one of the reasons that the paper was given the award …”, listen here:

Here is the original talk:

Algorithms

Live from NIPS 2017, presentations from the Algorithms session:• A Linear-Time Kernel Goodness-of-Fit Test• Generalization Properties of Learning with Random Features• Communication-Efficient Distributed Learning of Discrete Distributions• Optimistic posterior sampling for reinforcement learning: worst-case regret bounds• Regret Analysis for Continuous Dueling Bandit• Minimal Exploration in Structured Stochastic Bandits• Fast Rates for Bandit Optimization with Upper-Confidence Frank-Wolfe• Diving into the shallows: a computational perspective on large-scale shallow learning• Monte-Carlo Tree Search by Best Arm Identification• A framework for Multi-A(rmed)/B(andit) Testing with Online FDR Control• Parameter-Free Online Learning via Model Selection• Bregman Divergence for Stochastic Variance Reduction: Saddle-Point and Adversarial Prediction• Gaussian Quadrature for Kernel FeaturesLearning Linear Dynamical Systems via Spectral Filtering

Posted by Neural Information Processing Systems on Dienstag, 5. Dezember 2017

 

http://papers.nips.cc/paper/6630-a-linear-time-kernel-goodness-of-fit-test

In their paper the authors propose a novel adaptive test of goodness-of-fit, with computational cost linear in the number of samples. They learn the test features, which best indicates the differences between the observed samples and a reference model, by means of minimizing the false negative rate. These features are constructed via the Stein’s method, i.e. that it is not necessary to compute the normalising constant of the model. They further analyse the asymptotic Bahadur efficiency of the new test, and prove that under a mean-shift alternative, the test always has greater relative efficiency than a previous linear-time kernel test, regardless of the choice of parameters for that particular test. In experiments, the performance of their method exceeds that of the earlier linear-time test, and matches or exceeds the power of a quadratic-time kernel test. In high dimensions and where model structure may be exploited, this new goodness of fit test performs far better than a quadratic-time two-sample test based on the Maximum Mean Discrepancy, with samples drawn from the model.

The original paper can be downloaded via the NIPS pages:
https://nips.cc/Conferences/2017/Schedule?showEvent=8823

The paper is also available at arXiv:

Jitkrittum, W., Xu, W., Szabo, Z., Fukumizu, K. & Gretton, A. 2017. A Linear-Time Kernel Goodness-of-Fit Test. arXiv preprint arXiv:1705.07673.