German Constitutional Court Strikes Down Predictive Algorithms for Policing

Noting specially algorithms.  Meat to control a that level of specification?   Specifications of prediction? 

German Constitutional Court Strikes Down Predictive Algorithms for Policing

By Euractiv, February 17, 2023

Surveillance cameras at a German police station. 

In its ruling, the German Federal Constitutional Court struck down acts providing a statutory basis for police to process stored personal data through automated data analysis, in the case of Hesse, or automated data interpretation, in Hamburg.

The German Federal Constitutional Court declared the use of Palantir surveillance software by police in Hesse and Hamburg unconstitutional in a landmark ruling.

The ruling concludes a case brought by the German Society for Civil Rights (GFF) last year, hearings for which began in December. The plaintiffs argued that the software could be used for predictive policing, raising the risk of mistakes and discrimination by law enforcement. 

The German state of Hesse has been using the software since 2017, though it is not yet in place in Hamburg. The technology is provided by Palantir, a US data analytics firm which received early backing from intelligence agencies, including the CIA, FBI and NSA. 

The case was brought on behalf of 11 plaintiffs and rested on the argument that the software programme – named 'Hessendata' – facilitates predictive policing by using data to create profiles of suspects before any crime has been committed.

From Euractiv

View Full Article   

Predictive Database for Materials

Materials science becoming increasingly important

 Nanoengineers Develop Predictive Database for Materials

UC San Diego Today, Emerson Dameron, November 28, 2022

The M3GNet algorithm developed by nanoengineers at the University of California, San Diego (UCSD)'s Jacobs School of Engineering can forecast the structure and dynamic properties of any material almost instantaneously. Researchers used M3GNet to compile the matterverse.ai database of more than than 31 million yet-to-be-synthesized materials with traits predicted by machine learning algorithms. UCSD's Shyue Ping Ong and colleagues combined graph neural networks with many-body interactions into a highly accurate deep learning framework that operates across the entire periodic table. The team employed the Materials Project's database of materials energies, forces, and stresses to train the predictive M3GNet interatomic potential model. "We truly believe that the M3GNet architecture is a transformative tool that can greatly expand our ability to explore new material chemistries and structures," said Ong.

Full Article    

AI Model Transferability in Healthcare

 Key issues of how AI models make predictions in healthcare domains. 

ACM OPINION

AI Model Transferability in Healthcare: A Sociotechnical Perspective

By Nature Machine Intelligence, October 24, 2022

Doctor using a tablet to examine patient records.

To deliver value in healthcare AI and ML models must be integrated not only into technology platforms but also into local human and organizational ecosystems and workflows.

Predictive model transferability is gaining more attention as healthcare organizations attempt to implement artificial intelligence (AI)-based prediction tools. Although some machine learning (ML)-based models fail when subjected to retrospective validation across institutions and patient populations, technical improvements show promise for addressing this model efficacy problem. To address the engineering challenges, a technical subfield labelled MLOps has emerged.

However, the focus of MLOps on technical transferability may be obscuring a larger set of obstacles to sociotechnical transferability: organizational, social, and individual challenges of deploying models at scale across contexts, whether institutions, teams or individual roles....

To deliver value in healthcare AI and ML models must be integrated not only into technology platforms but also into local human and organizational ecosystems and workflows. ... 

From Nature Machine Intelligence

View Full Article    

NVIDIA Predicting Earth Quake Intensity

Predicting earthquake and their intensity was something we proposed some years ago, mentioned here.  glad to see the idea taken much further.    Looking at this application further. 

AI of Earthshaking Magnitude: DeepShake Predicts Quake Intensity   By Isha Salian

Tags: Deep Learning, featured, Geoscience, News

In a major earthquake, even a few seconds of advance warning can help people prepare — so Stanford University researchers have turned to deep learning to predict strong shaking and issue early alerts.

DeepShake, a spatiotemporal neural network trained on seismic recordings from around 30,000 earthquakes, analyzes seismic signals in real time. By observing the earliest detected waves from an earthquake, the neural network can predict ground shaking intensity and send alerts throughout the area. 

Geophysics and computer science researchers at Stanford used a university cluster of NVIDIA GPUs to develop the model, using data from the 2019 Ridgecrest sequence of earthquakes in Southern Califonia. 

When tested with seismic data from Ridgecrest’s 7.1 magnitude earthquake, DeepShake provided simulated alerts to nearby seismic stations 7 to 13 seconds before the arrival of high intensity ground shaking.

Most early warning systems pull multiple information sources, first determining the location and magnitude of an earthquake before calculating ground motion for a specific area. 

“Each of these steps can introduce error that can degrade the ground shaking forecast,” said Stanford student Daniel Wu, who presented the project at the 2021 Annual Meeting of the Seismological Society of America. 

Instead, the DeepShake network relies solely on seismic waveforms for its rapid early warning and forecasting system. The unsupervised neural network learned which features of seismic waveform data best forecast the strength of future shaking. 

“We’ve noticed from building other neural networks for use in seismology that they can learn all sorts of interesting things, and so they might not need the epicenter and magnitude of the earthquake to make a good forecast,” said Wu. “DeepShake is trained on a preselected network of seismic stations, so that the local characteristics of those stations become part of the training data.”  ... ' 

Predictive and Prescriptive Analytics Impacting the Bottom Line

Yes, well done, it always has.   And further methods like AI, a kind of analytics, does the same.  Been doing it for a lifetime.  Interesting numbers below.

How predictive and prescriptive analytics impact the bottom line  by 7wData

In a digitally transformed world, the combination of data and analytics is critical to maintaining a competitive advantage and business relevance. To achieve this goal, enterprises collect vast volumes of data and derive valuable insights from them. This knowledge can be anything from ascertaining customer satisfaction to identifying operational discrepancies.

The capability of business intelligence and analytics is continually evolving. In a highly competitive business world, analytics plays a key role in identifying trends and patterns to make quick and informed business decisions. Predictive and prescriptive analytics are two important methods in business-analytics solutions. Mordor Intelligence research suggests that the predictive and prescriptive analytics market (valued at $8.14 billion in 2019) is expected to grow at a compound annual rate of 22.53% to reach $27.57 billion by 2025.

As artificial intelligence (AI) and machine learning evolve and play a more significant role in data and analytics, smart algorithms can now pull both prescriptive and predictive insights from the data. Both approaches give insight and foresight to enable smart decision-making; they incorporate data mining, machine learning and statistical modeling to deliver deep insight into customers and overall operations ...."

Sewage as New Data Source

A long time promoter of finding new sources of data, especially where they can be predictive early warning indicators.   An argument for experimenting with data from new sources.  Here an example that shows the possibilities.

Cities are using sewer systems as COVID-19 early warning signs
New Haven, Connecticut and Carmel, Indiana both say sewage data is the canary in a coal mine  By Nicole Wetsman in TheVerge

Each day, workers at the wastewater treatment plant in New Haven, Connecticut, siphon off a bit of sewage and put it in a cooler. Then, researchers from Yale University swing by to pick it up. In their hands, that pile of refuse is a key tool to predict the trajectory of the local COVID-19 outbreak.

Cities around the United States are dipping into their sewer systems to track the levels of the novel coronavirus circulating inside their populations. If someone is infected with the virus, it shows up in their feces, even before they might feel sick. Virus-flecked feces make their way through sewage systems — and checking in on all that sewage gives public health officials another layer of data on the extent of the outbreak.

“It gives us a better idea of what’s going on in the city,” says New Haven epidemiologist Brian Weeks, who uses the data collected by the Yale team. .... "

Finding Future Buyers

Predicting future customers.  Or categories of users.

How IBM uses AI to help Harley-Davidson find future riders

Harley-Davidson also used IBM tech for its new electric LiveWire motorcycle.

Harley-Davidson is using artificial intelligence to find customers as it has struggled with a sales slump.

IBM has worked with the motorcycle maker on the AI project, according to Robert Thomas, IBM’s data and artificial intelligence general manager. He discussed it this week during an interview with FOX Business’ Lauren Simonetti on “Mornings with Maria” about the business applications of AI.

“The work we’ve done with Harley-Davidson —  they’re now making better predictions about future motorcycle owners,” Thomas said. “They could not do that manually — too much data, too much work that would have to occur.”    ... "

Will AI Save Lives?

More on the healthcare uses pf AI methods for accurate diagnosis:

It's too soon to tell if DeepMind's medical AI will save any lives
Artificial intelligence trained on health records can now detect kidney injury up to two days before it occurs. The idea is that an advance warning could help doctors intervene earlier to prevent irreversible damage to the kidneys.

AIs are already touted as rivals to doctors when it comes to detecting medical conditions such as certain cancers and childhood illnesses. But few undergo rigorous clinical trials, so it’s still too early to know whether they are effective in practice.

Nenad Tomašev at DeepMind and colleagues trained an algorithm to predict the likelihood that a person who was admitted to hospital would go on to develop acute kidney injury (AKI).

They trained the AI using de-identified electronic health records from 703,782 US veterans aged between 18 and 90, who were admitted to hospital between October 2011 and September 2015.

Read more: AIs that diagnose diseases are starting to assist and replace doctors

AKI results in a dramatic drop in the rate at which the kidneys filter blood. This causes a decrease in urine production and a build-up of waste products in the blood, such as creatinine, a by-product of muscle breakdown. Both of these are used as measures for diagnosis.

Based on creatinine levels from a patient’s medical records, at a given time point the AI predicted whether a kidney injury would occur within the next 48 hours. Its accuracy was confirmed by comparing the prediction to whether the patient was later diagnosed.

The algorithm was fairly accurate at predicting the most severe forms of AKI. It correctly predicted 90 per cent of the cases in which the patient’s kidney function deteriorated so severely that they eventually required long-term dialysis.

It is difficult for doctors to anticipate kidney injury, so that level of accuracy is significant given the consequences of a severe injury, which include death or the need for a kidney transplant, says Eric Topol at Scripps Research in the US, who was not involved in the research.

However, the algorithm was far less accurate for all forms of AKI, correctly predicting only 55.8 of all episodes, with a ratio of two false alerts for one correct prediction. .... "

Predicting Humor for Engagement

Humor as engagement.     Can it be delivered artificially, as a strong component of story?

It's No Joke: AI Beats Humans at Making You Laugh    |by Dina Gerdeman  in HBSWK

We all enjoy sharing jokes with friends, hoping a witty one might elicit a smile—or maybe even a belly laugh. Here’s one for you:

A lawyer opened the door of his BMW, when, suddenly, a car came along and hit the door, ripping it off completely. When the police arrived at the scene, the lawyer was complaining bitterly about the damage to his precious BMW.

"Officer, look what they've done to my Beeeeemer!” he whined.

"You lawyers are so materialistic, you make me sick!” retorted the officer. "You're so worried about your stupid BMW that you didn't even notice your left arm was ripped off!”

“Oh, my god,” replied the lawyer, finally noticing the bloody left shoulder where his arm once was. “Where's my Rolex?!”

Do you think your friends would find that joke amusing—well, maybe those who aren’t lawyers?

A research team led by Harvard Business School post-doctoral fellow Michael H. Yeomans put this laughing matter to the test. In a new study, he used that joke and 32 others to determine whether people or artificial intelligence (AI) could do a better job of predicting which jokes other people consider funny. ... "

Very Fast PhotoGrammetry for Agriculture

Once again, very fast acquisition of accurate maps, in agriculture could be used to quickly detect changes in plantings by location and predict changes.  Our own needs were to determine changes that would influence harvesting plans.  Another use for drones.

Army Photogrammetry Technique Makes 3D Aerial Maps in Minutes   By Devin Coldewey in TechCrunch

Researchers at the U.S. Army's Geospatial Research Laboratory in Virginia have developed a highly efficient photogrammetric method that can turn aerial imagery into accurate three-dimensional (3D) surface maps in near-real time without any human oversight. The Army’s 101st Airborne Division tested the system by flying a drone over Fort Campbell in Kentucky. The system was able to map a mock city used for training exercises. "Whether it's for soldiers or farmers, this tech delivers usable terrain and intelligence products fast," said Quinton King, a manager at TechLink, the Defense Department's commercial tech transfer organization.   ... "

Data Management for Data Science

In Kdnuggets a good description and visualization of data management needed for data science.

Everything a Data Scientist Should Know About Data Management

For full-stack data science mastery, you must understand data management along with all the bells and whistles of machine learning. This high-level overview is a road map for the history and current state of the expansive options for data storage and infrastructure solutions. By Phoebe Wong and Robert Bennett.

To be a real “full-stack” data scientist, or what many bloggers and employers call a “unicorn,” you have to master every step of the data science process — all the way from storing your data, to putting your finished product (typically a predictive model) in production. But the bulk of data science training focuses on machine/deep learning techniques; data management knowledge is often treated as an afterthought. Data science students usually learn modeling skills with processed and cleaned data in text files stored on their laptop, ignoring how the data sausage is made.   ... " 

Removing Bias from Predictive Modeling

Podcast of interest from Wharton re Bias, Podcast at the link.

Wharton's James Johndrow discusses his research on removing human bias from predictive modeling.

Predictive modeling is supposed to be neutral, a way to help remove personal prejudices from decision-making. But the algorithms are packed with the same biases that are built into the real-world data used to create them. Wharton statistics professor James Johndrow has developed a method to remove those biases. His latest research, “An Algorithm for Removing Sensitive Information: Application to Race-independent Recidivism Prediction,” focuses on removing information on race in data that predicts recidivism, but the method can be applied beyond the criminal justice system. He spoke to Knowledge@Wharton about his paper, which is co-authored with his wife, Kristian Lum, lead statistician with the Human Rights Data Analysis Group. (Listen to the podcast at the top of this page.)

An edited transcript of the conversation follows.

Knowledge@Wharton: Predictive modeling is becoming an increasingly popular way to assist human decision-makers, but it’s not perfect. What are some of the drawbacks?

James Johndrow: There has been a lot more attention lately about it, partly because things are being automated so much. There’s just more and more interest in having automatic scoring, automatic decision-making, or at least partly automatic decision-making. The area that I have been especially interested in — and this is a lot of work that I do with my wife — is criminal justice.  .... "

Simple Statistics Types in one Picture

A good, nontechnical infographic that describes commonly used simple statistics.  Just enough to use with non-technical management and decision makers.  Make sure to follow up with examples from your own data to make the usage clear.

Descriptive vs. Inferential Statistics in One Picture  From DSC
Posted by Stephanie Glen 

This simple picture shows the differences between descriptive statistics and Inferential statistics.  ... '

Nike Links RFID to Predictive Analytics for Inventory

Good example of sensors and analytics and transparency in the supply chain.

Nike to marry predictive analytics and RFID to optimize inventory performance
by Tom Ryan in Retailwire plus expert commentary.

Nike Inc. has acquired Celect, a predictive analytics firm founded by MIT professors, to accelerate its ability to match inventories to consumer needs.

Celect’s cloud-based analytics platform allows retailers to optimize inventory across an omnichannel environment through hyper-local demand predictions. Celect’s team will be integrated into Nike’s operations. Its co-founders will continue as tenured professors at MIT, consulting Nike on an ongoing basis.

“As demand for our product grows, we must be insight-driven, data optimized and hyper-focused on consumer behavior,” said Eric Sprunk, Nike’s COO, in a statement. “This is how we serve consumers more personally at scale.”

In a column for Retail Touchpoints from July, Andrea Morgan-Vandome, Celect’s chief marketing officer, wrote that advancements in artificial intelligence and machine learning now provide retailers with a more accurate view of demand across channels to choose the best fulfillment strategy based on product availability, likely demand, capacity constraints, shipping costs, delivery timing and other factors.

At the store level, such insights would reveal that a location seeing high inventory turnover wouldn’t be able to cover walk-in demand if it was also fulfilling online orders. Vice versa, a store seeing slower turnover risks becoming overstocked if it didn’t support online orders.

She wrote, “For each fulfillment decision that needs to be made, advanced optimization can account for the overall margin profitability and customer satisfaction by identifying the immediate payoff versus the long-term opportunity cost — instantly.”   .... ' 

Gartner Blog: We are Very early in Predictive Analytics

Here an excerpt.    Interesting thought, and our forecasting will become much better, in some areas spectacularly so.  But still, it will be operating with some of the same data. Old, faulty, in the wrong context, gathered haphazardly.  So you still cannot expect exact predictions.  If you are not exactly right there is the risk of implemented error.  And that risk itself cannot be perfectly tagged.   So there again a caution.   And I would certainly not use the position of a technology on a wavy line as predictive input data.  So please, save your pennies, but caution all around.

Start Saving for Predictive Analytics
by Steve Rietberg  |  August 8, 2019  | 

When I was young and fresh out of college, people urged me to start contributing to a retirement fund as soon as I could. They explained that since I had time on my side, even modest investments would pay significant dividends. I just needed the discipline to start saving early.

We’re all young, with respect to predictive analytics technology. According to Gartner’s Hype Cycle for CRM Sales, sales predictive analytics (which includes predictive forecasting, upsell/cross-sell recommendations and opportunity scoring) is still in its adolescence. This market is expected to grow quickly in the immediate future.

How can we take advantage of our youth, and get a head start on preparing for advances in predictive pipeline analytics?

Your CRM Thinks Too Linearly

Consider this. Your historic pipeline data suggests a correlation between sales stage and an opportunity’s likelihood to close. In many organizations, that correlation–directly or indirectly–informs seller coaching and forecasting decisions. But there is a disconnect between the sequential sales stages in a typical CRM system and the nonlinear buying process that customers actually follow. To improve the validity and accuracy of your pipeline analytics, you need to track more than sales stage. For this reason, measuring buyer behavior along with seller-provided sales stage and probability clears a path to improved predictive analytics. .... " 

Repairing a Satellite with Deep Learning AI in Space

By predicting lost data from other existing sources using deep learning.  Note the alternative uses of the approach, say in the case of solar storms.  Considerable complexity with varying goals.

IBM helped NASA fix one of its satellites using cutting-edge deep learning A.I.
How do you fix a satellite that’s floating 22,000 miles above the Earth’s surface?

That’s a question that NASA had to answer when it ran into problems with one of its crucial satellites. The satellite in question was the Solar Dynamics Observatory (SDO), which launched in 2010 with the important goal of studying the Sun and the effects of solar activity on Earth. This is important for all sorts of reasons — not least because solar storms can knock out GPS satellites, shut down electrical grids, and scramble communications.

Unfortunately, one of the SDO’s three instruments, responsible for measuring ultraviolet light, stopped working due to a fault. This data is essential to satellite operators, since it can affect the flight path of orbiting satellites. Not properly compensating for atmospheric changes due to ultraviolet light may cause satellites to fall out of orbit and burn up or crash.

It was deemed too costly to repair the $850 million satellite in space. As a result, NASA called in experts from IBM, SETI, Nimbix, Lockheed Martin, and its own Frontier Development Lab to see if they could solve the problem from Earth using cutting-edge artificial intelligence. The request? Could they figure out how to use data from the SDO’s remaining two instruments — its atmospheric imaging assembly and helioseismic and magnetic imager — to work out the missing ultraviolet radiation measurements. The answer: Apparently, yes.

“One of the biggest challenges was to find the optimal A.I. framework and model for the problem at hand — namely, virtually ‘resurrecting’ the failed SDO instrument so that we could once again get the data that instrument would have produced if it was still working,” Graham Mackintosh, A.I. advisor to SETI and NASA, told Digital Trends. “The team automated the task of modifying, testing, and recording the results of almost 1,000 different versions of the deep learning model before settling on the final approach they determined to be optimal.”  .... "

Operations and Predictive Analytics

Useful to think about operations.   Prediction implies we can adjust operations within a process based on future predictions.   Risk also come forward because of inaccuracy of predictions.

Predictive analytics in hybrid IT: The future of ops in TechBeacon   By David Linthicum, Chief Cloud Strategy Officer, Deloitte Consulting

The predictive analytics systems of today and tomorrow will change the way we do operations. We will know how system modifications will affect IT operations, security, and governance risks. We'll also learn how to automate forthcoming complexity in ways that are cheaper and less risky, and we will have the ability to proactively plan for three years into the future.

The growth of complexity in both on-premises and public cloud platforms, or in hybrid IT, is obvious to everyone at this point. Ops-related predictive analytics means the ability to leverage AI and big data in new, more efficient ways to deal with their increasing complexity. 

So, are you in? Most people in IT operations management, including cloud and traditional, see the value of systems that can literally predict the future. Apply that magic to IT Ops, and you have the ability to solve problems before they become known problems, perhaps problems that are never known to humans.

However, the costs of leveraging predictive analytics with ops are going up. This is true even with the use of the public cloud and its ability to leverage newer tooling and data sources. You'll need upgraded skill sets, expensive tooling replacements and upgrades, and, initially, more people at the helm.  ...  "

Scientific American on Deep Learning

Scientific American does a good job of providing a good,  intuitive and largely non-technical view of the math and applications of deep learning.   Good for use with management with a reasonable amount of patience.   I would not call it deep or complete, but enough for a useful intro.

A Deep Dive into Deep Learning
A personal journey to understand what lies beneath the startling powers of advanced neural networks
By Peter Bruce on April 10, 2019 in Sciam

On Wednesday, March 27, the 2018 Turing Award in computing was given to Yoshua Bengio, Geoffrey Hinton and Yann LeCun for their work on deep learning. Deep learning by complex neural networks lies behind the applications that are finally bringing artificial intelligence out of the realm of science fiction into reality. Voice recognition allows you to talk to your robot devices. Image recognition is the key to self-driving cars. But what, exactly, is deep learning?

Dozens of articles tell you that it’s a complex, multilayered neural network. But they don’t really shed much light on deep learning’s seemingly magical powers. For example, to explain how it can recognize faces out of a matrix of pixel values (i.e., an image).

As a data science educator, for years I have been seeking a clear and intuitive explanation of this transformative core of deep learning—the ability of the neural net to “discover” what machine learning specialists call “higher level features.” Older statistical modeling and machine learning algorithms, including neural nets, worked with databases where those features with predictive power already exist. In predicting possible bank failure, for example, we would guess that certain financial ratios (return on assets, return on equity, etc.) might have predictive value. In predicting insurance fraud, we might guess that policy age would be predictive.  .... " 

Digital Transformation

And I had just posted about Predictive maintenance.   Here elevator maker Thyssenkrupp AG talks about the technology and how it links to their own experience in digital transformation.   See my previous notes on Thyssenkrupp tag below.   Good piece here:

An elevator maker reveals the ups and downs of digital transformation  By Paul Gillin in SiliconAngle

A list of the 10 large companies that have most successfully transformed their businesses through technology published in the Harvard Business Review last year included many of the usual suspects – and one that many people probably wouldn’t even recognize.

Thyssenkrupp AG, a German materials giant that can trace its roots back more than 200 years, made the list on the strength of the 47 percent of its total sales that now come from new growth areas. .... " 

AI for Chemical Reaction Predictions

CSIG (Cognitive Systems Institute Group) Talk — Nov 15, 2018 - 10:30 AM ET
Talk Title: Artificial Intelligence for Chemical Reaction Predictions - IBM RXN for Chemistry 

Speaker: Dr. Teodoro Laino, IBM Research - Zurich   10:30-11am US Eastern 

Abstract: Organic synthesis is one of the key stumbling blocks in medicinal chemistry. A necessary yet unsolved step in planning synthesis is solving the forward problem: given  reactants and reagents, predict the products. We treat reaction prediction as a machine translation problem between SMILES strings of reactants-reagents and the products. We show  that a multi-head attention MolecularTransformer model outperforms all algorithms in the literature, achieving a top-I accuracy above 90% on a Common benchmark dataset. Our  algorithm requires no handcrafted rules, and accurately predicts subtle chemical transformations. Crucially, our model can accurately estimate its own uncertainty, with an uncertainty  score that is 89% accurate in terms of classifying whether a prediction is correct. I will present the underlying model as well as the free online platform for reaction predictions, named  IBM RXN for chemistry, http://rxn.res.ibm.com 

Bio: Teodoro Laino received his degree in theoretical chemistry in 2001 (Universityof Pisa and Scuola Normale Superiore di Pisa) and the doctorate in 2006 in computational  chemistry at the Scuola Normale Superiore di Pisa, Italy. His doctoral thesis, entitled "Multi-Grid QM/ MM Approaches in ab initio Molecular Dynamics" was supervised by Prof. Dr.  Michele Parrinello. From 2006 to 2008, he worked as a post-doctoral researcher in the research group of Prof. Dr. Jürg Hutter at the University of Zurich, where he developed  algorithms for ab initio and classical molecular dynamics simulations. Since 2008, he has been working in the department of Cognitive Computing and Industry Solutions at the IBM  Research - Zurich Laboratory (ZRL). The focus of his research is on complex molecular dynamics simulations for industrial-related problems (energy storage, life sciences and nano-  electronics) and on the application of machine learning/artificial intelligence technologies to chemistry and materials science problems. 

Zoom meeting Link: https://zoom.us/j/7371462221; Zoom Cailin: (415) 762-9988 or (646) 568-7788 Meeting id 7371462221 
Zoom International Numbers: https://zoom.us/zoomconference 
Check http://cognitive-science.info/community/weekly-update/for recordings & slides, and for any date & time changes 
Join Linkedin Group: https://www.linkedin.com/groups/6729452/ (Cognitive Systems Institute) to receive notifications 
Thu, Nov 15, 10:30am US Eastern https://zoom.us/j/7371462221 

More Details Here : http://cognitive-science.infb/commun;ty/weekly-update   (slides and recording will be posted here)    @sumalaikä 

 @teodorolaino of @IBMResearch on #ArtificialIntelligence for Chemical Reaction Predictions - IBM RXN for Chemistry in weekly talk series: cognitive-science.info/community/week…  #CSIGnews #opentechai #issip #MachineLearning  @KarolynSchalk @mattganis