---

Event Speakers

Albert King

Chief Data Officer, The Scottish Government
Head of Data & Digital Identity Division

Wayne Butterfield

Global Head of Intelligent Automation Solutions | Automation Expert
ISG Automation

Majella Clarke

AI Artist | Senior Data and AI Strategist
Finland-Australia Business Council

Igor Rotin

Chief Data Scientist / Head of Digital Lab
Liebherr Aerospace & Transportation

Natasha Ochiel

Co-Founder
AI Centre of Excellence, Kenya

Pedro Berrocoso

Global CoE Digital Innovation Lead
Takeda

Giselle Mota

Top 100 Future of Work Thought-Leader
Principal, Future of Work, ADP – Always Designing for People.

Johan Steyn

Author, AI & Automation Thought Leader
Chair, Special Interest Group – AI & Robotics, IITPSA (Institute of Information Technology Professionals South Africa)

Andeed Ma

Leader of AI Risk Chapter | Cognitive Technologies Thought-Leader
Risk and Insurance Management Association of Singapore

Sean Culey

Business transformation expert, award-winning keynote speaker and author of ‘Transition Point: From Steam to the Singularity’ | Client Partner – Aera Technology

Emma Ruttkamp-Bloem

AI Ethics Researcher Professor and Head of Department of Philosophy, University of Pretoria
AI Ethics Lead at South African Centre for AI Research (CAIR)

Franck Gazzola

Photographer & International Development, UNDER THE POLE
Executive Director & Owner, FROTHERS GALLERY

Raul Villamarin Rodriguez

Vice President, Woxsen University
Harvard Business Review, Advisory Council Member

Alexandra Ebert

Chief Trust Officer, MOSTLY AI | Chair, IEEE Synthetic Data IC Expert Group

Thomas Helfrich

CEO & Founder, instarel.ai
SwissCognitive AI Nerd | LinkedIn Influencer

Lee Coulter

Digital Transformation Leader | Operating Partner, Acresis LLC
Chair, Working Group on Standards in Intelligent Process Automation, IEEE

Yves Mulkers

Leading Business Intelligence & Data Architect
7wData, Founder

Andreas Welsh

Automation and AI Thought-Leader
Vice President Solution Management, Extensions & Artificial Intelligence, SAP

Payam Mokhtarian

Head of Data & AI
Plerion

---

Event Host & Facilitator

Dalith Steiger

Co-Founder, Global AI Thought Leader
SwissCognitive, World-Leading AI Network

Andy Fitze

Co-Founder, Digital Transformation Strategist
SwissCognitive, World-Leading AI Network

---

Event Team

Livia Spiesz

External Communications & Global Business Relations Lead
SwissCognitive, World-Leading AI Network

Bianka Németvölgyi

Social Media Manager
SwissCognitive, World-Leading AI Network

---

Virtual Conference

These virtual conferences are regular worldwide-reaching online events bringing dozens of global AI leaders and experts together to share their views, experiences and expertise in the development of AI to the benefit of business and society. These 3 hour-long events are transparently addressing the development of cognitive technologies – including successes and challenges – while reaching and connecting a global online community of over ½ million followers.

All the sessions and formats are strictly content-driven with a non-sales approach, allowing focused and open discussions with no BS just content. These events provide not only a platform to brainstorm and network but also to position experts, leaders, organisations, research developments, the current status and future outlook of AI. 

---

Der Beitrag How to set up an AI Centre of Excellence – 2nd Edition erschien zuerst auf SwissCognitive | AI Ventures, Advisory & Research.

Copyright by www.expresscomputer.in

Researchers at the University of Helsinki have developed a technique in which a computer models visual perception by monitoring human brain signals. In a way, it is as if the computer tries to imagine what a human is thinking about. As a result of this imagining, the computer is able to produce entirely new information, such as fictional images that were never before seen.

The technique is based on a novel brain-computer interface. Previously, similar brain-computer interfaces have been able to perform one-way communication from brain to computer, such as spell individual letters or move a cursor.

As far as is known, the new study is the first where both the computer’s presentation of the information and brain signals were modelled simultaneously using artificial intelligence methods. Images that matched the visual characteristics that participants were focusing on were generated through interaction between human brain responses and a generative neural network.

The study was published in the Scientific Reports journal in September. Scientific Reports is an online multidisciplinary, open-access journal from the publishers of Nature.

Neuroadaptive generative modelling
The researchers call this method neuroadaptive generative modelling. A total of 31 volunteers participated in a study that evaluated the effectiveness of the technique. Participants were shown hundreds of AI-generated images of diverse-looking people while their EEG was recorded.

The subjects were asked to concentrate on certain features, such as faces that looked old or were smiling. While looking at a rapidly presented series of face images, the EEGs of the subjects were fed to a neural network, which inferred whether any image was detected by the brain as matching what the subjects were looking for.

Based on this information, the neural network adapted its estimation as to what kind of faces people were thinking of. Finally, the images generated by the computer were evaluated by the participants and they nearly perfectly matched with the features the participants were thinking of. The accuracy of the experiment was 83 per cent.

“The technique combines natural human responses with the computer’s ability to create new information. In the experiment, the participants were only asked to look at the computer-generated images. The computer, in turn, modelled the images displayed and the human reaction toward the images by using human brain responses. From this, the computer can create an entirely new image that matches the user’s intention,” says Tuukka Ruotsalo, Academy of Finland Research Fellow at the University of Helsinki, Finland and Associate Professor at the University of Copenhagen, Denmark. […]

Reade more: www.expresscomputer.in

Der Beitrag A computer that can read and predict your thoughts erschien zuerst auf SwissCognitive | AI Ventures, Advisory & Research.

Copyright by Jürgen Schmidhuber 

1. The Decade of Long Short-Term Memory (LSTM)

Much of AI in the 2010s was about the NN called Long Short-Term Memory (LSTM) [LSTM1-13] [DL4]. The world is sequential by nature, and LSTM has revolutionized sequential data processing, e.g., speech recognition, machine translation, video recognition, connected handwriting recognition, robotics, video game playing, time series prediction, chat bots, healthcare applications, you name it. By 2019, [LSTM1] got more citations per year than any other computer science paper of the past millennium. Below I’ll list some of the most visible and historically most significant applications.

2009: Connected Handwriting Recognition. Enormous interest from industry was triggered right before the 2010s when out of the blue my PhD student Alex Graves won three connected handwriting competitions (French, Farsi, Arabic) at ICDAR 2009, the famous conference on document analysis and recognition. He used a combination of two methods developed in my research groups at TU Munich and the Swiss AI Lab IDSIA: LSTM (1990s-2005) [LSTM1-6] (which overcomes the famous vanishing gradient problem analyzed by my PhD student Sepp Hochreiter [VAN1] in 1991) and Connectionist Temporal Classification [CTC] (2006). CTC-trained LSTM was the first recurrent NN or RNN [MC43] [K56] to win any international contests.

CTC-Trained LSTM also was the First Superior End-to-End Neural Speech Recognizer. Already in 2007, our team successfully applied CTC-LSTM to speech [LSTM4], also with hierarchical LSTM stacks [LSTM14]. This was very different from previous hybrid methods since the late 1980s which combined NNs and traditional approaches such as Hidden Markov Models (HMMs), e.g., [BW] [BRI] [BOU] [HYB12]. Alex kept using CTC-LSTM as a postdoc in Toronto [LSTM8].

CTC-LSTM has had massive industrial impact. By 2015, it dramatically improved Google’s speech recognition [GSR15] [DL4]. This is now on almost every smartphone. By 2016, more than a quarter of the power of all those Tensor Processing Units in Google’s datacenters was used for LSTM (and 5% for convolutional NNs) [JOU17]. Google’s on-device speech recognition of 2019 (not any longer on the server) is still based on LSTM. See [MIR], Sec. 4. Microsoft, Baidu, Amazon, Samsung, Apple, and many other famous companies are using LSTM, too [DL4] [DL1].

2016: The First Superior End-to-End Neural Machine Translation was also Based on LSTM. Already in 2001, my PhD student Felix Gers showed that LSTM can learn languages unlearnable by traditional models such as HMMs [LSTM13]. That is, a neural “subsymbolic” model suddenly excelled at learning “symbolic” tasks! Compute still had to get 1000 times cheaper, but by 2016-17, both Google Translate [GT16] [WU] (which mentions LSTM over 50 times) and Facebook Translate [FB17] were based on two connected LSTMs [S2S], one for incoming texts, one for outgoing translations – much better than what existed before [DL4]. By 2017, Facebook’s users made 30 billion LSTM-based translations per week [FB17] [DL4]. Compare: the most popular youtube video (the song “Despacito”) got only 6 billion clicks in 2 years. See [MIR], Sec. 4.

LSTM-Based Robotics. By 2003, our team used LSTM for Reinforcement Learning (RL) and robotics, e.g., [LSTM-RL]. In the 2010s, combinations of RL and LSTM have become standard. For example, in 2018, an RL-trained LSTM was the core of OpenAI’s Dactyl which learned to control a dextrous robot hand without a teacher [OAI1].

2018-2019: LSTM for Video Games. In 2019, DeepMind beat a pro player in the game of Starcraft, which is harder than Chess or Go [DM2] in many ways, using Alphastar whose brain has a deep LSTM core trained by RL [DM3]. An RL-trained LSTM (with 84% of the model’s total parameter count) also was the core of OpenAI Five which learned to defeat human experts in the Dota 2 video game (2018) [OAI2] [OAI2a]. See [MIR], Sec. 4.

The 2010s saw many additional LSTM applications, e.g., [DL1]. LSTM was used for healthcare, chemistry, molecule design, lip reading [LIP1], stock market prediction, self-driving cars, mapping brain signals to speech (Nature, vol 568, 2019), predicting what’s going on in nuclear fusion reactors (same volume, p. 526), etc. There is not enough space to mention everything here.

2. The Decade of Feedforward Neural Networks

LSTM is an RNN that can in principle implement any program that runs on your laptop. The more limited feedforward NNs (FNNs) cannot (although they are good enough for board games such as Backgammon [T94] and Go [DM2] and Chess). That is, if we want to build an NN-based Artificial General Intelligence (AGI), then its underlying computational substrate must be something like an RNN. FNNs are fundamentally insufficient. RNNs relate to FNNs like general computers relate to mere calculators. Nevertheless, our Decade of Deep Learning was also about FNNs, as described next.

2010: Deep FNNs Don’t Need Unsupervised Pre-Training! In 2009, many thought that deep FNNs cannot learn much without unsupervised pre-training [MIR] [UN0-UN5]. But in 2010, our team with my postdoc Dan Ciresan [MLP1] showed that deep FNNs can be trained by plain backpropagation [BP1] (compare [BPA] [BPB] [BP2] [R7]) and do not at all require unsupervised pre-training for important applications. Our system set a new performance record [MLP1] on the back then famous and widely used image recognition benchmark called MNIST. This was achieved by greatly accelerating traditional FNNs on highly parallel graphics processing units called GPUs. A reviewer called this a “wake-up call to the machine learning community.” Today, very few commercial NN applications are still based on unsupervised pre-training (used in my first deep learner of 1991). See [MIR], Sec. 19.

2011: CNN-Based Computer Vision Revolution. Our team in Switzerland (Dan Ciresan et al.) greatly sped up the convolutional NNs (CNNs) invented and developed by others since the 1970s [CNN1-4]. The first superior award-winning CNN, often called “DanNet,” was created in 2011 [GPUCNN1,3,5]. It was a practical breakthrough. It was much deeper and faster than earlier GPU-accelerated CNNs [GPUCNN]. Already in 2011, it showed that deep learning worked far better than the existing state-of-the-art for recognizing objects in images. In fact, it won 4 important computer vision competitions in a row between May 15, 2011, and September 10, 2012 [GPUCNN5] before a similar GPU-accelerated CNN of Univ. Toronto won the ImageNet 2012 contest [GPUCNN4-5] [R6].

At IJCNN 2011 in Silicon Valley, DanNet blew away the competition through the first superhuman visual pattern recognition in a contest. Even the New York Times mentioned this. It was also the first deep CNN to win: a Chinese handwriting contest (ICDAR 2011), an image segmentation contest (ISBI, May 2012), a contest on object detection in large images (ICPR, 10 Sept 2012), at the same time a medical imaging contest on cancer detection. (All before ImageNet 2012 [GPUCNN4-5] [R6].) Our CNN image scanners were 1000 times faster than previous methods [SCAN], with tremendous importance for health care etc. Today IBM, Siemens, Google and many startups are pursuing this approach. Much of modern computer vision is extending the work of 2011, e.g., [MIR], Sec. 19.

Already in 2010, we introduced our deep and fast GPU-based NNs to Arcelor Mittal, the world’s largest steel maker, and were able to greatly improve steel defect detection through CNNs [ST] (before ImageNet 2012). This may have been the first Deep Learning breakthrough in heavy industry, and helped to jump-start our company NNAISENSE. The early 2010s saw several other applications of our Deep Learning methods.

Through my students Rupesh Kumar Srivastava and Klaus Greff, the LSTM principle also led to our Highway Networks [HW1] of May 2015, the first working very deep FNNs with hundreds of layers. Microsoft’s popular ResNets [HW2] (which won the ImageNet 2015 contest) are a special case thereof. The earlier Highway Nets perform roughly as well as ResNets on ImageNet [HW3]. Highway layers are also often used for natural language processing, where the simpler residual layers do not work as well [HW3].

3. LSTMs & FNNs, especially CNNs. LSTMs v FNNs

In the recent Decade of Deep Learning, the recognition of static patterns (such as images) was mostly driven by CNNs (which are FNNs; see Sec. 2), while sequence processing (such as speech, text, etc.) was mostly driven by LSTMs (which are RNNs [MC43] [K56]; see Sec. 1). Often CNNs and LSTMs were combined, e.g., for video recognition. FNNs and LSTMs also invaded each other’s territories on occasion. Two examples:

1. Multi-dimensional LSTM [LSTM15] does not suffer from the limited fixed patch size of CNNs and excels at certain computer vision problems [LSTM16]. Nevertheless, most of computer vision is still based on CNNs.

2. Towards the end of the decade, despite their limited time windows, FNN-based Transformers [TR1] [TR2] started to excel at Natural Language Processing, a traditional LSTM domain (see Sec. 1). Nevertheless, there are still many language tasks that LSTM can rapidly learn to solve quickly [LSTM13] (in time proportional to sentence length) while plain Transformers can’t.

Business Week called LSTM “arguably the most commercial AI achievement” [AV1]. As mentioned above, by 2019, [LSTM1] got more citations per year than all other computer science papers of the past millennium [R5]. The record holder of the new millennium [HW2] is an FNN related to LSTM: ResNet [HW2] (Dec 2015) is a special case of our Highway Net (May 2015) [HW1], the FNN version of vanilla LSTM [LSTM2].

4. GANs: the Decade’s Most Famous Application of our Curiosity Principle (1990)

Another concept that has become very popular in the 2010s are Generative Adversarial Networks (GANs), e.g., [GAN0] (2010) [GAN1] (2014). GANs are an instance of my popular adversarial curiosity principle from 1990 [AC90, AC90b] (see also survey [AC09]). This principle works as follows. One NN probabilistically generates outputs, another NN sees those outputs and predicts environmental reactions to them. Using gradient descent, the predictor NN minimizes its error, while the generator NN tries to make outputs that maximize this error. One net’s loss is the other net’s gain. GANs are a special case of this where the environment simply returns 1 or 0 depending on whether the generator’s output is in a given set [AC19]. (Other early adversarial machine learning settings [S59] [H90] neither involved unsupervised NNs nor were about modeling data nor used gradient descent [AC19].) Compare [SLG] [R2] [AC18] and [MIR], Sec. 5.

5. Other Hot Topics of the 2010s: Deep Reinforcement Learning, Meta-Learning, World Models, Distilling NNs, Neural Architecture Search, Attention Learning, Fast Weights, Self-Invented Problems …

In July 2013, our Compressed Network Search [CO2] was the first deep learning model to successfully learn control policies directly from high-dimensional sensory input (video) using deep reinforcement learning (RL) (see survey in Sec. 6 of [DL1]), without any unsupervised pre-training (extending earlier work on large NNs with compact codes, e.g., [KO0] [KO2]; compare more recent work [WAV1] [OAI3]). This also helped to jump-start our company NNAISENSE.

A few months later, neuroevolution-based RL (see survey [K96]) also successfully learned to play Atari games [H13]. Soon afterwards, the company DeepMind also had a Deep RL system for high-dimensional sensory input [DM1] [DM2]. See [MIR], Sec. 8.

By 2016, DeepMind had a famous superhuman Go player [DM4]. The company was founded in 2010, by some counts the decade’s first year. The first DeepMinders with AI publications and PhDs in computer science came from my lab: a co-founder and employee nr. 1.

Our work since 1990 on RL and planning based on a combination of two RNNs called the controller and the world model [PLAN2-6] also has become popular in the 2010s. See [MIR], Sec. 11. (The decade’s end also saw a very simple yet novel approach to the old problem of RL [UDRL].)

For decades, few have cared for our work on meta-learning or learning to learn since 1987, e.g., [META1] [FASTMETA1-3] [R3]. In the 2010s, meta-learning has finally become a hot topic [META10] [META17]. Similar for our work since 1990 on Artificial Curiosity & Creativity [MIR] (Sec. 5, Sec. 6) [AC90-AC10] and Self-Invented Problems [MIR] (Sec. 12) in POWERPLAY style (2011) [PP] [PP1] [PP2]. See, e.g., [AC18].

Similar for our work since 1990 on Hierarchical RL, e.g. [HRL0] [HRL1] [HRL2] [HRL4] (see [MIR], Sec. 10), Deterministic Policy Gradients [AC90], e.g., [DPG] [DDPG] (see [MIR], Sec. 14), and Synthetic Gradients [NAN1-NAN4], e.g., [NAN5] (see [MIR], Sec. 15).

Similar for our work since 1991 on encoding data by factorial disentangled representations through adversarial NNs [PM2] [PM1] and other methods [LOC] (compare [IG] and [MIR], Sec. 7), and on end-to-end-differentiable systems that learn by gradient descent to quickly manipulate NNs with Fast Weights [FAST0-FAST3a] [R4], separating storage and control like in traditional computers, but in a fully neural way (rather than in a hybrid fashion [PDA1] [PDA2] [DNC] [DNC2]). See [MIR], Sec. 8.

Similar for our work since 2009 on Neural Architecture Search for LSTM-like architectures that outperform vanilla LSTM in certain applications [LSTM7], e.g., [NAS], and our work since 1991 on compressing or distilling NNs into other NNs [UN0] [UN1], e.g., [DIST2] [R4]. See [MIR], Sec. 2.

Already in the early 1990s, we had both of the now common types of neural sequential attention: end-to-end-differentiable “soft” attention (in latent space) [FAST2] through multiplicative units within networks [DEEP1-2] (1965), and “hard” attention (in observation space) in the context of RL [ATT0] [ATT1]. This led to lots of follow-up work. In the 2010s, many have used sequential attention-learning NNs. See [MIR], Sec. 9.

Many other concepts of the previous millennium [DL1] [MIR] had to wait for the much faster computers of the 2010s to become popular.

As mentioned in Sec. 21 of ref [MIR], surveys from the Anglosphere do not always make clear [DLC] that Deep Learning was invented where English is not an official language. It started in 1965 in the Ukraine (back then the USSR) with the first nets of arbitrary depth that really learned [DEEP1-2] [R8]. Five years later, modern backpropagation was published “next door” in Finland (1970) [BP1]. The basic deep convolutional NN architecture (now widely used) was invented in the 1970s in Japan [CNN1], where NNs with convolutions were later (1987) also combined with “weight sharing” and backpropagation [CNN1a]. We are standing on the shoulders of these authors and many others – see 888 references in ref [DL1].

Of course, Deep Learning is just a small part of AI, in most applications limited to passive pattern recognition. We view it as a by-product of our research on more general artificial intelligence, which includes optimal universal learning machines such as the Gödel machine (2003-), asymptotically optimal search for programs running on general purpose computers such as RNNs, etc.

6. The Future of Data Markets and Privacy

AIs are trained by data. If it is true that data is the new oil, then it should have a price, just like oil. In the 2010s, the major surveillance platforms (e.g., Sec. 1) [SV1] did not offer you any money for your data and the resulting loss of privacy. The 2020s, however, will see attempts at creating efficient data markets to figure out your data’s true financial value through the interplay between supply and demand. Even some of the sensitive medical data will not be priced by governmental regulators but by patients (and healthy persons) who own it and who may sell parts thereof as micro-entrepreneurs in a healthcare data market [SR18] [CNNTV2].

Are surveillance and loss of privacy inevitable consequences of increasingly complex societies? Super-organisms such as cities and states and companies consist of numerous people, just like people consist of numerous cells. These cells enjoy little privacy. They are constantly monitored by specialized “police cells” and “border guard cells”: Are you a cancer cell? Are you an external intruder, a pathogen? Individual cells sacrifice their freedom for the benefits of being part of a multicellular organism.

Similar for super-organisms such as nations [FATV]. Over 5000 years ago, writing enabled recorded history and thus became its inaugural and most important invention. Its initial purpose, however, was to facilitate surveillance, to track citizens and their tax payments. The more complex a super-organism, the more comprehensive its collection of information about its components.

200 years ago, at least the parish priest in each village knew everything about all the village people, even about those who did not confess, because they appeared in the confessions of others. Also, everyone soon knew about the stranger who had entered the village, because some occasionally peered out of the window, and what they saw got around. Such control mechanisms were temporarily lost through anonymization in rapidly growing cities, but are now returning with the help of new surveillance devices such as smartphones as part of digital nervous systems that tell companies and governments a lot about billions of users [SV1] [SV2]. Cameras and drones [DR16] etc. are becoming tinier all the time and ubiquitous; excellent recognition of faces and gaits etc. is becoming cheaper and cheaper, and soon many will use it to identify others anywhere on earth – the big wide world will not offer any more privacy than the local village. Is this good or bad? Anyway, some nations may find it easier than others to become more complex kinds of super-organisms at the expense of the privacy rights of their constituents [FATV].

7. Outlook: 2010s v 2020s – Virtual AI v Real AI?

In the 2010s, AI excelled in virtual worlds, e.g., in video games, board games, and especially on the major WWW platforms (Sec. 1). Most AI profits were in marketing. Passive pattern recognition through NNs helped some of the most valuable companies such as Amazon & Alibaba & Google & Facebook & Tencent to keep you longer on their platforms, to predict which items you might be interested in, to make you click at tailored ads etc. However, marketing is just a tiny part of the world economy. What will the next decade bring?

In the 2020s, Active AI will more and more invade the real world, driving industrial processes and machines and robots, a bit like in the movies. (Better self-driving cars [CAR1] will be part of this, especially fleets of simple electric cars with small & cheap batteries [CAR2].) Although the real world is much more complex than virtual worlds, and less forgiving, the coming wave of “Real World AI” or simply “Real AI” will be much bigger than the previous AI wave, because it will affect all of production, and thus a much bigger part of the economy. That’s why NNAISENSE is all about Real AI.

Some claim that big platform companies with lots of data from many users will dominate AI. That’s absurd. How does a baby learn to become intelligent? Not “by downloading lots of data from Facebook” [NAT2]. No, it learns by actively creating its own data through its own self-invented experiments with toys etc, learning to predict the consequences of its actions, and using this predictive model of physics and the world to become a better and better planner and problem solver [AC90] [PLAN2-6].

We already know how to build AIs that also learn a bit like babies, using what I have called artificial curiosity since 1990 [AC90-AC10] [PP-PP2], and incorporating mechanisms that aid in reasoning [FAST3a] [DNC] [DNC2] and in the extraction of abstract objects from raw data [UN1] [OBJ1-3]. In the not too distant future, this will help to create what I have called in interviews see-and-do robotics: quickly teach an NN to control a complex robot with many degrees of freedom to execute complex tasks, such as assembling a smartphone, solely by visual demonstration, and by talking to it, without touching or otherwise directly guiding the robot – a bit like we’d teach a kid [FA18]. This will revolutionize many aspects of our civilization.

Sure, such AIs have military applications, too. But although an AI arms race seems inevitable [SPE17], almost all of AI research in the 2020s will be about making human lives longer & healthier & easier & happier [SR18]. Our motto is: AI For All! AI won’t be controlled by a few big companies or governments. Since 1941, every 5 years, compute has been getting 10 times cheaper [ACM16]. This trend won’t break anytime soon. Everybody will own cheap but powerful AIs improving her/his life in many ways.

So much for now on the 2020s. In the more distant future, most self-driven & self-replicating & curious & creative & conscious AIs [INV16] will go where most of the physical resources are, eventually colonizing and transforming the entire visible universe [ACM16] [SA17] [FA15] [SP16], which may be just one of countably many computable universes [ALL1-3].

Der Beitrag The 2010s: Our Decade of Deep Learning / Outlook on the 2020s erschien zuerst auf SwissCognitive | AI Ventures, Advisory & Research.

Copyright by Persönlich and ABB
Interview Reiner Schönrock 

As the computer scientist John McCarthy posited in 1956, “every aspect … of intelligence can in principle be so precisely described that a machine can be made to simulate it”.

Indeed, only 30 years after this fundamental principle of artificial intelligence (AI) was formulated, the first chess computers shocked the world. Today, another 30 years later, we tell our car where to go, dictate pages of correspondence to a word processing program and complain to a computer-generated hotline assistant.

These systems translate human speech into binary code and, if desired, back again –  into any language. This allows the Bulgarian taxi driver to understand where the German tourist in the backseat would like to go.

Such tremendous computing power is contained within a handy smartphone that can also capture high definition photos, stream movies and shows from vast libraries, and enable instantaneous international video conferencing. Today, nearly every other person on the planet has access to the mostly free services of this “mobile intelligence”.

During a train journey along Lake Zurich, we talked to Dalith Steiger, co-founder of the award-winning organization SwissCognitive – The Global AI Hub (www.swisscognitive.ch). Steiger is one of the world’s leading influencers in the area of AI and was recently named a Top 100 Digital Shaper by Bilanz magazine.

SwissCognitive has more than 400,000 followers on social media, where it posts updates on the latest developments in AI.

As an expert in the field, Steiger is convinced AI will change almost every aspect of our lives. Since the conversation took place on a train, the topic of mobility soon came up.

Steiger: “The world grows more complex every day. Tremendous strain is being put on infrastructure, tech is engaged in a constant race with itself, climate change is forcing us to rethink our ways and many societal norms are changing. These factors demand and facilitate new solutions to practices that for centuries were ‘just the way we do things’. The way we approach mobility, for instance, will change dramatically. Change will soon be the constant among topics such as electromobility, the sharing economy, traffic density and smart cities.”

Experts agree that in the long term the gradual introduction of AI is the only way to ensure the widespread breakthrough of pioneering options such as electromobility. AI allows communication between mobile and fixed elements in the value-added chain, a critical component in making these processes truly practical. When a motorist drains the battery in their electric car after 400 km, they don’t want to wait two hours for a charging station to become available – they want to dock immediately at a station that has anticipated their arrival. ABB charging stations, which are managed and maintained on a remote network, collect the necessary data. In the near future, AI solutions could make this network more robust to meet wider demand.

What else is AI changing in mobility? Will we soon be driving remote-controlled, autonomous vehicles?

That’s doubtful, says Steiger: “People like to think in extremes, and many are already
imagining an age without drivers. At SwissCognitive, we believe that priority will be given to environmental work and managing freeway capacity. It makes more economic and ecological sense to increase use on some sections of the highway than to spend lots of money on expanding them. Preferably, this would be accomplished by gradually introducing semi-autonomous trucks or even passenger vehicles. These projects are not intended to force out drivers; instead, the goal is to increase infrastructural capacity. This approach could also be applied to rail and air travel.”

Although semi-autonomous cars, buses and trucks could be on the road in the foreseeable future, there is a lot human drivers can do that these vehicles cannot: they cannot care for, maintain or repair themselves – a flat tire will literally throw them off course.

The entire process – from design to manufacture to distribution – will present challenges to even the most advanced systems for a long time to come. In short, for the time being AI-supported systems can manage clearly defined autonomous tasks – no more, no less.

Steiger: “We refer to systems today as ‘narrow AI’. They can handle only a single problem at a time. These systems still struggle to solve more complex problems, such as those in which moving images, written text and spoken language must be analyzed and recognized in context.”

At ABB Future Labs, however, technology for the AI-based factories of the future is already being developed. Eventually, autonomous industry systems will not only be able to compile and analyze data from different sources, but reach independent conclusions based on that information. They will thus be in a position to make correct decisions, even in situations they have not been programmed to handle.

ABB has already taken the first steps towards this future; for instance, recently an unmanned ferry was directed through Helsinki harbor by remote control. In the autonomous shipping of the future, a single captain could monitor several such ships from land, intervening only when necessary.

More than 1,000 ships and their technical components are already monitored by the nine ABB Ability Collaboration Operations Centers around the globe. This allows companies to anticipate maintenance requirements and have the necessary replacement parts ready when the ship comes into port. It also enables route optimization, which benefits the environment by lowering energy use and CO2 emissions, improves passenger comfort and protects cargo.

Steiger on the environmental aspects of AI:

“Today, we are facing a growing need to use resources more effectively. This, coupled with the abundance of data collected by the Internet of Things (IoT), is opening a host of new possibilities for AI. Smart technology represents a real chance to attain the UN’s Sustainable Development Goals by its proposed deadline. This applies to education, the distribution of critical medicines and human rights as much as it does to fighting climate change.”

Furthermore: “To take advantage of these opportunities, however, society needs to be open to new approaches. Regulators, like the rest of us, have not always kept an eye on big advances in technology. What we’re asking of regulators in finance, communications, aviation, pharmaceuticals or transportation is to allow us greater leeway in developing new ideas. We are a country of doers, and doers need space to test the viability of their investments. Test first, regulate second – that should be the rule.”

Read the complete interview.

Der Beitrag On course for a sustainable future with artificial intelligence erschien zuerst auf SwissCognitive | AI Ventures, Advisory & Research.

copyright by thenextweb.com

Helsinki-based Noiseless Acoustics and Amsterdam-based OneWatt are relying on artificial intelligence (AI) to better understand the sound patterns of troubled machines. Through AI they are enabling faster and easier problem detection. Both companies were also finalists in last year’s New Energy Challenge , an initiative by Shell, YES!Delft, and Rockstart that looks at innovative technologies and solutions within European and Israeli startups for the energy transition. Following last year’s success, a 2018 edition of the challenge was just launched.

The value of noiseless sound

According to the U.S. Department of Energy, industrial motor use accounts for 25 percent of all electricity usage nationwide. Yet despite the vital roles they play, motors can fail for any number of reasons, leading to a loss in productivity and profitability.

But what if it was possible to transform that noiseless sound into value? Through the use of AI, sounds can be analyzed to detect machine failure. In other words: Making sound visible even when it can’t be heard. With the aid of non-invasive sensors, machine learning algorithms, and predictive maintenance solutions, failing components can be recognized at an early stage before they become a major issue.

OneWatt is preventing problems by listening to motors. Through its Embedded Acoustic Recognition Sensors (EARS) device, combined with machine learning and frequency analysis, OneWatt can detect and predict faults before they happen. This includes the what, when, and where of a problem.

16,000 sound clips of faulty motors

The startup used its device among the top eight motor faults in the industry. These ranged from bearing faults to soft footing faults. By doing so the company collected almost 2TB of acoustic data containing over 16,000 sound clips of faulty motors.

“Audio is the most apparent sign of mechanical failure,” Paolo Samontañez, CTO of OneWatt, told me in an interview. “Most of the faults are signaled in this domain because of the movement of the components in the motor creating friction. Visible light is not a good indicator since it is not able to see through the motor, and could not tell if the bearings are degrading.”

Ultrasound is an option to visualize the internals of the motor, but Samontañez says this is costly. It would also require an operator to move the transmitter and receiver around, similar to an ultrasound machine in a hospital. Audio is the ideal solution, mainly because it’s unobtrusive. This also happens to be a primary requirement when dealing with industrial facilities as they need assurance that there will be no negative effect on the motors when a device is installed. […]

read more – copyright by thenextweb.com

Der Beitrag AI can now ‘listen’ to machines to tell if they’re breaking down erschien zuerst auf SwissCognitive | AI Ventures, Advisory & Research.