Upcoming ISSIP/CSG Talk: Robots in the Pandemic

Correspondent Jim Spohrer talks about robot tech in our pandemic futures.    I will post about this and point to the transcript.

ISSIP Speaker Series: COVID-19 & Future of Work and Learning

Speaker: Jim Spohrer, Director, Cognitive Open Tech, IBM
Title: How will COVID-19 affect the need for and use of robots in a service world with less physical contact?
Date & Time: June 24, 2020, 12:30-1:00 PM US Pacific Time, (on Zoom, info below)

Abstract: As AI and robotics come to the service world, including retail, hospitality, education, healthcare, and government, some jobs will go away, some new jobs will be created, and the income required for a family to thrive might be lessened.   In this creative session participants will be asked to engage in discussing three scenarios below – and the wicked problem of the bespoke impact on livelihood and jobs, which is creating uncertainty and concerns.   The groups will then report back on which scenarios they find more desirable. Click here for more details about this session.

More on this talk and background:  http://www.issip.org/about-issip/community/covid-19-working-group/

Recorded talk, slides:   https://youtu.be/RchxIKum_tI 

See also:  http://www.issip.org/   The International Society of Service Innovation Professionals, ISSIP 

ISSIP Newsletter: https://mailchi.mp/2f401b893caa/issip-june-2020-newsletter?e=78b83a31fb

(Update) Talk Oct 31: On the Need of an Ethical AI Due Diligence

CSIG:   Cognitive Systems Group Talk
Speaker: Roberto Zicari, Goethe University, Frankfurt Germany
Talk:       On the need of an ethical AI Due Diligence
Time:     October 31, 2019    11:30 ET

Invitation to the ISSIP Cognitive Systems Institute Group Webinar

Please join us for the next ISSIP and  CSIG Speaker Series (see details below, or click here).

Roberto Zicari (Professor of Database and Information Systems (DBIS) - Goethe University Frankfurt) on the need of an Ethical AI Due Diligence
When:  Thursday, October 31, 11:30am - US Eastern - (4:30pm Germany time)

Background :

Roberto V. Zicari is professor of Database and Information Systems (DBIS) at the Goethe University Frankfurt, Germany. He is an internationally recognized expert in the field of Databases and Big Data. His interests also expands to Ai and Ethics, Innovation and Entrepreneurship. , He is the founder of the Frankfurt Big Data Lab at the Goethe University Frankfurt, and the editor of the ODBMS.org web portal and of the ODBMS Industry Watch Blog. He was for the past five years a visiting professor with the Center for Entrepreneurship and Technology within the Department of Industrial Engineering and Operations Research at UC Berkeley (USA). 

Task Description : AI is becoming a sophisticated tool in the hands of a variety of stakeholders, including political leaders. Some AI applications may raise new ethical and legal questions, and in general have a significant impact on society (for the good or for the bad or for both). People motivation plays a key role here. , With AI the important question is how to avoid that it goes out of control, and how to understand how decisions are made and what are the consequences for society at large. , In this talk, Roberto V. Zicari will present some preliminary thoughts on the concept of an AI Ethical Inspection Process. This could be part of an Ethics by Design process, or if the AI has already been designed, it can be used to do an ethical sanity check, so that a certain AI Ethical standard of care is achieved. It can be used by a variety of AI stakeholders.

Slides:  http://cognitive-science.info/wp-content/uploads/2019/10/CSIGTalkZicari.20191031.pdf
Talk:    https://youtu.be/jrwuZvt_H7k

Zoom meeting Link: https://zoom.us/j/7371462221
Zoom Callin: (415) 762-9988 or (646) 568-7788 Meeting id 7371462221
Zoom International Numbers: https://zoom.us/zoomconference
Thu, Oct 31, 11:30am -12:00am US Eastern   @     https://zoom.us/j/7371462221
More Details and recordings will be posted Here : http://cognitive-science.info/community/weekly-update/

Sponsored by our Cognitive Systems Institute and ISSIP (International Society of Service Innovation Professionals)

Please pass this on to people with interest in AI, ethics.

Talk: Sports Summary Highlight Video Construction Using AI

You can now find the recording from Stephen Hammer's excellent Oct 24th talk on improving the fan experience:    "Sports Summary Highlight Videos using AI"

Slides:    http://cognitive-science.info/wp-content/uploads/2019/10/csig_Sports__AI_Hammer_v1.120191024-2.pdf
Talk:      https://youtu.be/StDgf3mnKEU

#CSIGnews #opentechai #ISSIP #Wimbledon @usta #GRAMMYs @BaughmanAaron  @IBMWatsonMedia

Via Susan Malaika

Upcoming and past talks, given most weeks:   http://cognitive-science.info/community/weekly-update/ 

Standardizing Deep Learning Model Deployment

Got this more detailed invite late,  and the talk is over,  but below you can follow along on slides and talk, see below:

Nick Pentreath, IBM : "Standardizing deep learning model deployment with the Model Asset Exchange"   Technical talk

Background :

Nick Pentreath is a principal engineer in IBM's Center for Open-source Data & AI Technology (CODAIT), where he works on machine learning. Previously, he cofounded Graphflow, a machine learning startup focused on recommendations. He has also worked at Goldman Sachs, Cognitive Match, and Mxit. He is a committer and PMC member of the Apache Spark project and author of "Machine Learning with Spark". Nick is passionate about combining commercial focus with machine learning and cutting-edge technology to build intelligent systems that learn from data to add business value.

Task Description : The popular version of applying deep learning is that you take an open-source or research model and simply deploy it. However, in reality developers and data scientists face many challenges ranging from custom requirements for data pre- and post-processing, to inconsistencies across frameworks, to lack of standardization in serving APIs. The goal of the IBM Developer Model Asset eXchange (MAX) is to remove these barriers to entry for developers to obtain, train and deploy open-source deep learning models for their enterprise applications. For model deployment, MAX provides container-based, fully self-contained model artefacts, encompassing the end-to-end deep learning predictive pipeline and exposing a standardized REST API. This talk explores the MAX deployment and serving framework, covering best practices for cross-framework, standardized deep learning model deployment  .... 

Slides and recording of the talk now posted:   http://cognitive-science.info/community/weekly-update/ 

Please retweet: https://twitter.com/sumalaika/status/1181291829091864581?s=20

Join LinkedIn Group https://www.linkedin.com/groups/6729452

What is Watson? A Significant Part of the Future of AI

Thoughtful piece.   Despite recent criticism as not having attained enough, Watson is a genuine accomplishment.    We saw it early on and remarked about how it was the direction for enterprise AI, solving difficult, data rich problems.   I continue to watch how IBM is advancing this.   See our ISSIP and CSI speaker series, which includes many IBMers who are making real development progress, much of it available to all.    Have worked with a number of IBM Watson practitioners.  As mentioned here, its a series of microservices that will become a key part of the future of AI.   Easy to explore today.

This Is Watson
IBM Watson: Reflections and Projections
Written by: Rob Thomas
Categorized: AI | Analytics | Data Science | IBM Watson | This Is Watson

AI has gone through many cycles since we first coined the term “machine learning” in 1959. Our latest resurgence began in 2011 when we put Watson on national television to play Jeopardy! against humans. This became a cornerstone event, demonstrating that we had something unique. And we saw early success, putting Watson to work on projects with clients. This created even more excitement. That excitement led to more opportunity. At this stage, we have a large product organization, separate dedicated research organization, and an entire health organization all leveraging and building on this technology.

So, what is Watson? This is the question I’ve been asked the most since IBM combined its Data and AI software units earlier this year.

Let’s start with what it’s not. Watson is not a personal assistant like Alexa, Siri or Google Assistant – its capabilities far exceed those of a consumer AI device. However, consumers likely interact with some form of Watson every day, they just are not aware of it. That’s because Watson was built to enable business-to-business interactions. Watson technology spans everything from powering virtual assistants to embedding AI in business processes across many industries.

Watson does not have a voice, gender or personality. Many people associate Watson with the measured male voice used to bring it to life on Jeopardy! and in older TV commercials. We gave it a voice for those instances, but it is not a box that talks back to you. It is a set of composable microservices (software) that live in the cloud. Any cloud, public or private.

Put simply, Watson is software capable of making sense of data sets and understanding natural language to provide recommendations, make predictions, and automate work. As we have fine-tuned our approach, the “Watson” name is only used on IBM products and solutions that significantly utilize IBM Watson technology. For products and solutions in which AI is an embedded enhancement, we use the designation “with Watson.”  .... " 

Standardizing Deep Learning Model Deployment

Tomorrow of interest:

Join @MLnick on Oct 10,  8:30am US Eastern (note unusual time) -  Nick Pentreath:  Principal Engineer, CODAIT (Center for Open-Source Data & AI Technologies) at IBM.    "Standardizing deep learning model deployment with the Model Asset Exchange" - in series cognitive-  http://cognitive-science.info/community/weekly-update/  (Slides and recordings)

#CSIGnews #opentechai #ISSIP - Check: https://t.co/SiHycih1il  @KarolynSchalk @ibrahimatlinux @mattganis

Join the meetings by pointing your web browser to:  https://zoom.us/j/7371462221 ; Callin: (415) 762-9988 or (646) 568-7788 Meeting id 7371462221 ; International Numbers: https://zoom.us/zoomconference

Join the CSIG LinkedIn Group to get reminders about talks and discuss them. Use twitter: #CSIGNews & #OpenTechAI

Advancing Data-enabled Research via Reproducibility and Transparency

Talk of interest I was unable to get to last week:

Cognitive Systems Institute Group Speaker Series, Sept 26, 2019.
By Victoria Stodden   

Advancing Data-enabled Research via Reproducibility and Transparency

Slides:  http://stanford.edu/~vcs/talks/CSIG2019-STODDEN.pdf
Talk:    https://youtu.be/GrG-2amslhk

1. Three Types of Reproducibility
2. Four Key Recommendations: “Reproducibility and
Replication in Science” National Academies of
Science, Engineering, and Medicine Report (2019)
3. Proposal: A Computable Scholarly Record .... '

Adversarial Robustness Toolbox

The below talk has already occured, the recording and slides will be placed here:  http://cognitive-science.info/community/weekly-update/   (The slides are there already)  Some very fundamental things here.

 ISSIP CSIG Speaker Series

Invitation to the ISSIP Cognitive Systems Institute Group Webinar
Please join us for the next ISSIP CSIG Speaker Series (see details below, or click here).
Beat Buesser, IBM , "The First Major Release of Adversarial Robustness 360 Toolbox (ART) v1.0 - A Milestone in AI Security "

Background :
Beat Buesser is a Research Staff Member of IBM Research in the AI & Machine Learning Group at the Dublin Research Laboratory. He is leading the development of the Adversarial Robustness 360 Toolbox (ART) and his research focuses on the security of machine learning and artificial intelligence. Before joining IBM, he worked as postdoctoral associate at the Massachusetts Institute of Technology and obtained his doctorate degree from ETH Zurich.

Task Description :  Adversarial Robustness 360 Toolbox (ART) is a Python library supporting developers and researchers in defending, certifying and verifying Machine Learning models against adversarial threats and helps making AI systems more secure and trustworthy. ART addresses growing concerns about people’s trust in AI, specifically the security of AI in mission-critical applications. In this talk we are presenting ART v1.0 which extends ART to non-neural-network models including gradient boosted decision trees, support vector machines (SVM), random forests, logistic regression, Gaussian processes, decision trees, Scikit-learn pipelines, black-box classifiers, etc. and extends the input data types beyond images to include tabular data and texts.

Join LinkedIn Group https://www.linkedin.com/groups/6729452

IBM AI Open Source Tool Explainability Talk

Upcoming talk, looks to be quite interesting regarding AI explain-ability open source method.  The talk will be recorded and I will post its location afterwards.

 CSIG (Cognitive Systems Institute Group) Talk - Thursday Sep 5, 2019 - 10:30-11am US Eastern
Title: Al Explainability 360 Toolkit

Speakers: Vijay Arya & Amit Dhurandhar, IBM Research

As AI and ML algorithms make inroads into society, calls are increasing for algorithms to explain their outputs. Affected citizens, government regulators. domain experts. or system developers. present different requirements for explanations. To address these needs we introduce:

AI Explainability 360 (http://aix360.mybluemix.net/)  (good tutorials there) , an open-source software toolkit featuring 8 state-of-the-art explainability methods and 2 evaluation metrics. We provide a taxonomy to help entities require explanations to navigate the space of explanation methods, in the toolkit and in the broader literature.

We have implemented an extensible software architecture that organizes methods according to their place in the AI modeling pipeline. We discuss enhancements to bring research innovations closer to consumers of explanations. ranging from algorithms. to tutorials and an interactive web demo to introduce AI explainability to different and application domains. Together, the toolkit and taxonomy can help identify gaps where more are needed and provide a platform to incorporate them as they are developed. 

Zoom meeting Link: https://zoom.us/j/7371462221

Zoom Callin: (415) 762-9988 or (646) 568-7788 Meeting id 7371462221
Zoom International Numbers: https://zoom.us/zoomconference
Thu, Aug 2, 10:30am US Eastern https://zoom.us/j/7371462221
More Details and recording Here : http://cognitive-science.info/community/weekly-update/

Dialogue Mapping Talk

Will be attending.

CSIG (Cognitive Systems Institute Group) Talk - July 25, 2019 -10:30-11am US Eastern  

Talk Title: Dialogue Mapping with IBIS for More Productive Meetings  Speaker: Paul Fernhout, Software Developer 

Abstract: Tens of billions of US dollars a year are wasted on unproductive unfun meetings. Worse, even "productive" meetings sometimes fail to consider a diversity of opinions and so produce suboptimal decisions (e.g. Fukushima Daiichi's seawall height). How can Cognitive Systems help people make better decisions in meetings more quickly?  How can we help people with strong disagreements collaborate in mapping landscape of possibilities in a fun way?  One option is to visualize the thinking going on in a meeting using Dialog Mapping(TM) developed by Jeff Conklin and associates, which visualizes discussions using the Issue-Based Information Systems (IBIS) grammar consisting  of Issues/Questions, Options/Answers, and supporting Pros & Cons. This talk will explain more about Dialogue Mapping and (hopefully) provide a live demonstration. 

Bio: Paul Fernhout is passionate about helping people collaborate to make better decisions more quickly using computers. He has worked  as a software developer on decision-support projects for a wide variety of organizations ranging from non-profits to multi-nationals to  governments, as well as on independent FOSS projects with his wife related to educational simulations, evolutionary design tools,  information organizers, and Participative Narrative Inquiry. He has also written about technology and social change. 

Zoom meeting Link: https://zoom.us/j/7371462221; Zoom Call in: (415) 762-9988 or (646) 568-7788 Meeting id 7371462221 

Zoom International Numbers: https://zoom.us/zoomconference 
http://cognitive-science.info/community/weekly-update/ for recordings & slides, and for any date & time changes 

Join Group: https://www.linkedin.com/groups/6729452/  (CognitiveSystemesInstitute)to receive notifications Thu, July 25, 10:30am US Eastern https://zoom.us/j/7371462221 
More Details Here : http://cognitive-science.info/community/weekly-update/

Via Karolyn Schalk,    Susan Malaika

CSI Talk: Teaching Data Science

From last week's CSIG Talk, 

Speaker:  Dr. Mine Cetinkaya-Rundel, Duke University:
Suppose our goal is to educate the new generation of data scientists working on machine learning and artificial intelligence problems, and especially those who are not intimidated by learning new computing technologies. Where do we start their education at the college level? Which topics do we cover in their first course, and which topics do we postpone till later? In this talk, we propose an introductory data science course that places a heavy emphasis on exploratory data analysis and modeling as well as collaboration, effective communication of findings, and ethical considerations as a welcoming and horizon broadening introduction to the discipline at large.

Mine Çetinkaya-Rundel is the Director of Undergraduate Studies and Associate Professor of the Practice in the Department of Statistical Science at Duke University as well as Data Scientist and Professional Educator at RStudio. She is also the creator and maintainer of datasciencebox.org and she teaches the popular Statistics with R MOOC on Coursera as well as numerous courses on DataCamp. ... 

Data Science in a Box: http://datasciencebox.org     

Good inclusion of data visualization, exploratory analysis, decision making, cautions to bias ...

Slides from talk.
Recording of talk.

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

Introduction to the Community Data License Agreement

CSIG (Cognitive Systems Institute Group) Talk — Thursday Nov 8, 2018 - 10:30-11am US Eastern 

Talk Title: "Introduction to the Community Data License Agreement: An "Open Source" Agreement Specifically Designed for Data and Content Sharing and Analysis in the Big Data world" 

Speaker: Christopher O'Neill, IBM 

Abstract: This talk will provide a general introduction the Community Data License Agreement ("CDLA") family of agreements, published  by The Linux Foundation in late 2017. Topics will include the general structure of the agreements, with a particular focus on their unique  data analysis terms and other aspects in which they are designed to address the unique characteristics of data from an Intellectual  Property perspective. 

Bio: Christopher O'Neill is Associate General Counsel — Intellectual Property Law at IBM Corporation, based in Armonk, New York. In over  25 years at IBM, Mr. O'Neill has held a variety of positions, both in IBM's product businesses and in its litigation group. In his current role,  Mr. O'Neill has responsibility for a variety of matters, including IP indemnity matters, data rights issues, adversely-held patent matters,  and open source issues. He graduated from New York University School of Law in 1987 and is a member of the New York bar. 

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.Iinkedin.com/groups/6729452/ (Cognitive Systems Institute) to receive notifications 
Thu, Nov 8, 10:30am US Eastern • https://zoom.us/j/7371462221 
More Details Here : http://cognitive-science.info/community/weekly-update/   (Also slides and talk recording will be placed here) 

More at: https://www.linuxfoundation.org/press-release/2017/10/linux-foundation-debuts-community-data-license-agreement/  #CSIGnews #opentechai #AI @KarolynSchalk @mattganis @jwaup @MishiChoudhary @t_streinz @hyurko

AI Community Day Presentations

Recording of the October 11 meetings.   Note you do have to sign up to see these, and as I understand it they will only be available for about 90 days.   Good resource provided via the Cognitive Systems Institute:

 ....  “Listen to more AI Community Day” More Info: https://www.ibmai-platform.bemyapp.com/#/conferences   (please sign-up first to see recordings) IBM, Figure Eight, Stack Overflow, Udacity, etc ... "

Talk: Improving Online Conversations

Interesting application idea.   Improving conversation is a great idea.  But I have often seen related approaches to essentially define 'good', as agreeing with me.   Will this be able to temper interaction to make it meaningful, or will conversational AI go the way of 'correcting' my speech?

CSIG (Cognitive Systems Institute Group)  

Talk Title: Improving On-Line Conversations through Machine Learning 
Speaker: Marie Pellat, Google   Oct 4, 2018 - 10:30-11am US Eastern 

Abstract: Having a healthy discussion online is difficult. The sad reality of internet today is that people are harassed into suicide,  journalists are threatened into silence, hate speech is normalized and sometimes organized. Approximately 1/4 of women face sexual  harassment online and are physically threatened. Over the last few years, many organizations have simply turned off their comments  sections as they became to hard to moderate. How can technology help? This talk will cover how the Conversation-Ai team at Google has  used deep learning techniques to help improve conversations online. We will also discuss the limitations of machine learning algorithms with respect to unintended bias and go over some mitigation techniques. 

Bio: Marie grew up in Paris and completed a BS in Mathematics and Physics at the Ecole Polytechnique. She later moved to California to complete a MS at Stanford University. After a couple years working on the ML team at Nest, Marie transferred to another Google bet called Jigsaw where she works on language models using artificial intelligence techniques and with the goal of improving conversations  online. 

Zoom meeting Link: https://zoom.us/j/7371462221 
Zoom Callin: (415) 762-9988 or (646) 568-7788 Meeting id 7371462221 
Zoom International Numbers: https://zoom.us/zoomconference 
Thu, Oct 4, 10:30am US Eastern https://zoom.us/j/7371462221 
More Details Here : http://cognitive-science.info/community/weekly-update/    (Slides and recording here) 

Bias and Fairness in Machine Learning

From the CSIG talk given today:

An instructive experiment which was released for use and experimentation today by IBM.  The slides instructive by themselves are here.   The complete audio and video of the presentation will be placed here shortly.   The comments on the presentation also point to other work that has been done and other efforts underway.   Based on the complexity of the problem there is some doubt that a universal solution to this problem is easily determined, give also the broad regulatory and even philosophical underpinning .  Also this is about Machine learning trained problems, not necessarily human decision making.    Still it would be useful to detect if some artifact of ML, like sampling is involved.  Also the need for integration of clear explanatory capabilities were mentioned.  The examples shown were still too technically complex for typical decision makers.

Nicely done.   Well worth examining.   I understand anyone can experiment with this, instructions in the talk. 

Talk:  “AI Fairness 360”
Speaker: Kush Varshney, IBM

Talk Description:  
Machine learning models are increasingly used to inform high stakes decisions about people. Although machine learning, by its very nature, is always a form of statistical discrimination, the discrimination becomes objectionable when it places certain privileged groups at systematic advantage and certain unprivileged groups at systematic disadvantage. Biases in training data, due to either prejudice in labels or under-/over-sampling, yields models with unwanted bias.In this presentation, we introduce AI Fairness 360, a new Python package that includes a comprehensive set of metrics for datasets and models to test for biases, explanations for these metrics, and algorithms to mitigate bias in datasets and models. They have developed the package with extensibility in mind.  They encourage the contribution of your metrics, explainers, and debiasing algorithms. Please join the community to get started as a contributor. ... " 

Talk: AI Fairness 360: Python Package

How to look at biases when your making AI driven decisions about people.

Invitation to the ISSIP Cognitive Systems Institute Group Webinar

 Slides and Recording: http://cognitive-science.info/community/weekly-update/

Date and Time: September 20, 2018 - 10:30am US Eastern
Talk Title: AI Fairness 360

Speaker: Kush Varshney, IBM
Talk Description:  
Machine learning models are increasingly used to inform high stakes decisions about people. Although machine learning, by its very nature, is always a form of statistical discrimination, the discrimination becomes objectionable when it places certain privileged groups at systematic advantage and certain unprivileged groups at systematic disadvantage. Biases in training data, due to either prejudice in labels or under-/over-sampling, yields models with unwanted bias.In this presentation, we introduce AI Fairness 360, a new Python package that includes a comprehensive set of metrics for datasets and models to test for biases, explanations for these metrics, and algorithms to mitigate bias in datasets and models. They have developed the package with extensibility in mind.  They encourage the contribution of your metrics, explainers, and debiasing algorithms. Please join the community to get started as a contributor.

Bio: 
Kush R. Varshney was born in Syracuse, NY in 1982. He received the B.S. degree (magna cum laude) in electrical and computer engineering with honors from Cornell University, Ithaca, NY, in 2004. He received the S.M. degree in 2006 and the Ph.D. degree in 2010, both in electrical engineering and computer science from the Massachusetts Institute of Technology (MIT), Cambridge.  While at MIT, he was a National Science Foundation Graduate Research Fellow.Dr. Varshney is a principal research staff member and manager with IBM Research AI at the Thomas J. Watson Research Center, Yorktown Heights, NY, where he leads the Learning and Decision Making group.  He is the founding co-director of the IBM Science for Social Good initiative.  He applies data science and predictive analytics to human capital management, healthcare, olfaction, computational creativity, public affairs, international development, and algorithmic fairness, which has led to recognitions such as the 2013 Gerstner Award for Client Excellence for contributions to the WellPoint team and the Extraordinary IBM Research Technical Accomplishment for contributions to workforce innovation and enterprise transformation. He conducts academic research on the theory and methods of statistical signal processing and machine learning. His work has been recognized through best paper awards at the Fusion 2009, SOLI 2013, KDD 2014, and SDM 2015 conferences. He is a senior member of the IEEE and a member of the Partnership on AI's Safety-Critical AI working group.

Date and Time : September 20 2018 - 10:30am US Eastern
Zoom meeting Link: https://zoom.us/j/7371462221
Zoom Callin: (415) 762-9988 or (646) 568-7788 Meeting id 7371462221
Zoom International Numbers: https://zoom.us/zoomconference
(Check the website in case the date or time changes: http://cognitive-science.info/community/weekly-update/ )

Please retweet  -

Join LinkedIn Group https://www.linkedin.com/groups/6729452

AI Overhyped? The Term is.

The problem is the term AI itself.   The assumption that it is far more than it is.   This does not mean you should not think about how smarter capabilities could be inserted into codes to augment our capabilities.  For a while we were using the term 'Cognitive Systems' to indicate methods closer and even mimicking human perception and abilities.  Probably be better to ditch 'AI' and go with Cognitive.  Though even the latter requires too much explanation and can be over emphasized.   Our Cognitive Systems Institute, monitored here,  attempts to emphasize cognitive aspects. Beware over-marketing.

 Artificial intelligence is often overhyped—and here’s why that’s dangerous

AI has huge potential to transform our lives, but the term itself is being abused in very worrying ways, says Zachary Lipton, an assistant professor at Carnegie Mellon University.
by Martin Giles

To those with long memories, the hype surrounding artificial intelligence is becoming ever more reminiscent of the dot-com boom.

Billions of dollars are being invested into AI startups and AI projects at giant companies. The trouble, says Zachary Lipton, is that the opportunity is being overshadowed by opportunists making overblown claims about the technology’s capabilities.

During a talk at MIT Technology Review’s EmTech conference today, Lipton warned that the hype is blinding people to its limitations. “It’s getting harder and harder to distinguish what’s a real advance and what is snake oil,” he said.

AI technology known as deep learning has proved very powerful at performing tasks like image recognition and voice translation, and it’s now helping to power everything from self-driving cars to translation apps on smartphones,

But the technology still has significant limitations. Many deep-learning models only work well when fed vast amounts of data, and they often struggle to adapt to fast-changing real-world conditions.

In his presentation, Lipton also highlighted the tendency of AI boosters to claim human-like capabilities for the technology. The risk is that the AI bubble will lead people to place too much faith in algorithms governing things like autonomous vehicles and clinical diagnoses.

“Policymakers don’t read the scientific literature,” warned Lipton, “but they do read the clickbait that goes around.” The media business, he says, is complicit here because it’s not doing a good enough job of distinguishing between real advances in the field and PR fluff.

Lipton isn’t the only academic sounding the alarm: in a recent blog post, “Artificial Intelligence—The Revolution Hasn’t Happened Yet,” Michael Jordan, a professor at University of California, Berkeley, says that AI is all too often bandied about as “an intellectual wildcard,” and this makes it harder to think critically about the technology’s potential impact. ... " 

Talk on Advances in Image Recognition

I note this is an advanced technical talk on elements of image recognition ...

Invitation to the ISSIP Cognitive Systems Institute Group Webinar

Full series list, past and present  recordings are here:  http://cognitive-science.info/community/weekly-update/

Date and Time: September 06, 2018 - 10:30am US Eastern
Talk Title: Learning to Find Good Correspondences

Speaker: Eduard Trulls, EPFL

Talk Description:  

In this talk, Eduart will present a novel deep architecture to learn to find good correspondences for wide-baseline stereo. Our solution is based on putative keypointmatches, which we learn to label as inliers or outliers while simultaneously using them to recover the camera pose, a fundamental Computer Vision problem. Our solution is simple (no convolutional or fully-connected layers), small (4 Mb), easy to train (state of the art matching outdoors scenes with only 59 images) and generalizes well, particularly in contrast to dense networks which require the entire image.

Bio: 
Eduard Trulls is currently a post-doc at the Computer Vision Lab at EPFL in Lausanne, Switzerland. He obtained his PhD from the Institute of Robotics in Barcelona, Spain, in 2015. His thesis explored novel strategies to enhance local, low-level features (e.g. SIFT, HOG) with global, mid-level data such as motion or segmentation cues. His current work focuses on designing novel approaches to apply deep learning techniques to classical, low-level computer vision problems such as local feature extraction and matching for 3D reconstruction.  ... 

Date and Time : September 06 2018 - 10:30am US Eastern
Zoom meeting Link: https://zoom.us/j/7371462221
Zoom Callin: (415) 762-9988 or (646) 568-7788 Meeting id 7371462221
Zoom International Numbers: https://zoom.us/zoomconference
(Check the website in case the date or time changes: http://cognitive-science.info/community/weekly-update/ )

Please retweet  - https://twitter.com/sumalaika/status/1036933287640489984

Join LinkedIn Group https://www.linkedin.com/groups/6729452

Requirements for an Enterprise AI Benchmark

(Update to a recent talk of note given on 8/23/2018)

Slides
Talk recording.

ISSIP Cognitive Systems Institute Group Webinar

full series here http://cognitive-science.info/community/weekly-update/

Talk Title: Requirements for an Enterprise AI Benchmark

Speakers: Cedric Bourrasset, Atos Bull; Rajesh Bordawekar, IBM

Talk Description:  

At present, AI benchmarks either focus on evaluating deep learning approaches or infrastructure capabilities. These approaches don’t capture end-to-end performance behavior of enterprise AI workloads. It is also clear that there is not one reference metric that will be suitable for all AI applications nor all existing platforms. Cedric and Rajesh first present the state of the art regarding the current basic and most popular AI benchmarks. They then present the main characteristics of AI workloads from various industrial domains. Finally, they focus on the needs for ongoing and future industry AI benchmarks and conclude on the gaps to improve AI benchmarks for enterprise workloads.  

Cedric Bourasset : After receiving a Ph.D. in Electronics and computer vision in 2016 from the Blaise Pascal University of Clermont-Ferrand defending the dataflow model of computation for FPGA High Level Synthesis problematic in embedded machine learning application, Cedric is now working as AI Product Manager at Atos Bull with the mission to develop Atos AI product line. One product is a software solution for developing AI enterprise solutions and the other one is computer vision solution for people detection, tracking and reidentification into multi-camera environments.

Rajesh Bordawekar:  Rajesh is a member of the Systems Acceleration department at the IBM T. J. Watson Research Center. Prior to joining IBM Research in September 1998, he was a post-doctoral fellow at the Center for Advanced Computing Research, California Institute of Technology. 
He received his PhD in Computer Engineering from Syracuse University. 
Rajesh studies interactions between applications, programming languages/runtime systems, and computer architectures.    He is interested in understanding how modern hardware, multi-core processors, GPUs, and SSDs impact design of optimal algorithms for main memory and out-of-core problems. .... 

Join LinkedIn Group https://www.linkedin.com/groups/6729452