Dorsey Learning Hall 1106, Luddy Hall
Gaurav Chopra / Assistant Professor – Analytical and Physical Chemistry
Engineering drug discovery using chemical data science
Abstract: Current drug discovery has not matched the accelerating rate of technology development observed in many other walks of life, getting exponentially more expensive (costing billions of dollars) and less efficient (taking a decade or more of time). We have developed automated methods that will accelerate several areas of the drug design pipeline from synthesis to bioactivity. Our approach employs deep learning compatible molecular representations (features) that reduce the time and cost of developing new molecules and processes while increasing their efficacy (desired property) because molecules (and related process) will be designed for specific properties rather than created using empirical knowledge. We will briefly discuss molecular representations and computational pipelines to develop a library of synthetically feasible bioactive molecules, and models for reactivity that will be used to predict and validate synthetic conditions for chemical reactions. We will also show a few examples of this combined model-based machine learning and experimental approach to select designs for function-specific diverse potent and non-toxic molecules for cancer and immune cell types tested in vivo mouse models and in pet dogs’ clinical study. Finally, we plan to bring all these models together in a virtual reality-based drug design game that we have developed to crowdsource automated molecular discovery of molecules with desired properties. We expect that engineering drug design based on phenotype (property) of interest will save cost and time by accelerating different aspects of the drug discovery processes.
Biography: Gaurav Chopra is an Assistant Professor in the Department of Chemistry at Purdue University and a member of the Purdue Center for Cancer Research and the Purdue Institutes of Data Science, Drug Discovery, Integrative Neuroscience, and Immunology. His laboratory brings together four very different fields of science, namely, chemistry, immunology, machine learning, and computer science to study chemical environments from atomic to molecular to cellular scales with a theme of model based chemistry. Chopra obtained his Ph.D. in computational structural chemistry/biology with Dr. Michael Levitt (2013 Chemistry Nobel Laureate) at Stanford University developing proteome scale methods for drug discovery. As a JDRF fellow, he trained in experimental immunology with Dr. Jeffrey A. Bluestone (University of California, San Francisco) to understand and chemically perturb the role of regulatory T-cells. Chopra is interested in developing analytical and computational methods to understand and chemically perturbing immune microenvironments with an application to combat disease by developing cell-specific synthetic molecules by degrading protein networks and by chemically engineering live cells that can also be used as drugs. His lab develops proteome-scale modeling methods, uses machine learning, immunological, and synthetic chemistry methods to discover, design, synthesize and verify chemical entities/tools that guide experiments done in his laboratory for studying immune cell subtypes (such as myeloid-derived suppressor cells and microglia). Chopra and collaborators in schools of medicine worldwide have used this approach to identify putative drug leads (new synthetic leads, combination of existing drugs) using in vitro and in vivo preclinical studies for more than 10 different diseases including cancer, immunological, metabolic, infectious and genetic indications (castration resistant prostate, invasive bladder cancer, type 1 diabetes, dental caries, dengue, herpes, drug resistant tuberculosis, etc). He has published a range of papers in PNAS, Cell-Immunity, Drug Discovery Today, Journal of Clinical Investigation and Immunology, Bioinformatics among others.
Chopra is passionate about innovation in the area of education and engaging the public for molecular visualization and discovery. His group has developed a virtual reality environment toolbox for molecular exploration, MINT (Molecular Interactions using New Technologies), as an educational and scientific research “game” where the chemistry principles of molecular interactions are taught using a dynamic immersive environment guided by his group’s software suite, CANDIY (Chemical Algorithms for Network based Decisions on Interactions for modeling reactivitY). The goal of MINT game is two-fold: discover chemistry concepts using a gaming environment to make it fun, and, to engage citizen scientists by crowdsourcing this game with a vision to enhance computational chemistry algorithms.
The laboratory’s GitHub repository is available at – https://github.com/chopralab
Dorsey Learning Hall 1106, Luddy Hall
Bledi Taska /Chief Economist / Burning Glass Technologies
Unlocking the Power of Big Data to Inform Labor Market and Education Policies
The pace at which technology is transforming jobs and skills is unprecedented. Automation, together with other disrupting sources such as skill hybridization or the gig economy, have led to a severe misalignment of the education and training system and the labor market.
Leading researchers like Tom Mitchell and Erik Brynjolfsson warn that “without data on how artificial intelligence is affecting jobs, policymakers will fly blind into the next industrial revolution.” Traditional survey data lacks the granularity to help us understand the future of work and education, and comes with significant lags and update costs. Real-time labor market data can crucially fill these gaps.
In this talk, Dr. Bledi Taska will provide a summary of real-time labor market data and discuss how they have been used by different researchers to answer important questions around labor market and education mismatch. Additionally, there will be a discussion of how different advanced predictive analytics models can be built upon this type of big data to better understand the future of work and education.
Dorsey Learning Hall 1106, Luddy Hall
Daniel H. Robertson, Ph.D., Director, Center for Applied Data Sciences, Indiana Biosciences Research Institute
Driving innovation between academic and industrial organizations showcasing the toxicogenomics collaborations
Abstract: IBRI is an new applied research institute in Indiana that was created to bridge the gap between academic and industrial organizations. The Center for Applied Data Sciences is one of the core centers of the institute with a focus on data access, data cleaning and integration, and novel analytics to attack challenging life sciences problems. This talk will provide a brief overview of the IBRI, 16-Tech (the future home of the IBRI), the Center for Applied Data Sciences and its collaborations. A specific case study, the Collaborative Toxicogenomics Platform will be reviewed in more depth. This collaboration is developing a shared platform to deliver harmonized analysis methods to identify and quantify potential compound toxicity from transcriptomic data. The scientific goals of this collaboration are to expose current industrial experience and decision-making tools, enable multiple standard technologies to be run consistently, and to move to an evidenced-based biological understanding of transcriptomic data. The platform will provide a shared collaborative platform for multiple users, leverage open-source tools and methodologies, and ultimately placed in the public domain for community-based future improvements.
Bio: Daniel H. Robertson, a proven and experienced technical leader in information technology (IT), computational science and research, is focused on defining and developing the IBRI’s computational analytics, digital, and data science capabilities. Dr. Robertson originally joined the IBRI in mid-2015 as part of a loaned executive program at Eli Lilly and Company, but in mid-2017 he accepted a permanent position at the IBRI to drive innovative research among multiple life sciences companies, academic institutions, and technology companies to advance solutions to critical problems.
His most recent role at Eli Lilly and Company was Senior Director of Research IT where he led the IT team supporting discovery systems and processes across six global research sites and nine functional/therapeutic areas. In 2000, he joined Lilly as a research scientist in Lilly Research Laboratories and performed numerous independent contributor and scientific leadership roles before transitioning to the IT organization in 2010. Prior to joining Lilly, he held multiple positions at IUPUI from 1993 through 2000, including Associate Scientist and Director of Technical and Administrative Services and Manager of the Facility for Computational Molecular Science, and as Lead Developer of VR.
Dr. Robertson earned his PhD in physical chemistry from Florida State University and his Bachelor of Science degree in chemistry, graduating Summa Cum Laude, from Florida Southern College. After earning his PhD., Dr. Robertson served as an NRC/NRL Postdoctoral Research Associate at the Naval Research Laboratory in Washington, D.C., Dr. Robertson has published 67 papers in refereed journals, authored three invited book chapters, and conducted more than 80 professional/technical presentations. He has been honored with multiple awards from Eli Lilly and Company and IUPUI, and he is a member of the American Chemical Society, American Physical Society and physics and mathematics honoraries.
Dorsey Learning Hall 1106, Luddy Hall
Fran Berman | Edward P. Hamilton Distinguished Professor of Computer Science - Rensselaer Polytechnic Institute
Civilizing the Internet of Things
Abstract: The ubiquity of digital information has ushered in a new environment that is transforming how we work, do business, and live: The “Internet of Things” (IoT) is an interconnected cyber-physical-biological environment that links devices, systems, data, and people. At its best, the IoT has the potential to create a rich environment that empowers people through technology, and technology through intelligence. At its worst, the IoT can open a Pandora’s Box of inappropriate behavior, unintended consequences, and intrusiveness. To create an IoT that advances society, and not just technology, involves a holistic approach that coordinates technological design and capability, responsible social and ethical policy and regulation, and environmental sustainability. In this talk, Fran Berman discusses challenges and opportunities in developing the IoT from wild west to empowering environment that promotes the public good.
Bio: Dr. Francine Berman is the Edward P. Hamilton Distinguished Professor in Computer Science at Rensselaer Polytechnic Institute. She is a Fellow of the Association of Computing Machinery (ACM) and a Fellow of the IEEE. In 2009, Dr. Berman was the inaugural recipient of the ACM/IEEE-CS Ken Kennedy Award for “influential leadership in the design, development, and deployment of national-scale cyberinfrastructure.” In 2015, Dr. Berman was nominated by President Obama and confirmed by the U.S. Senate to become a member of the National Council on the Humanities.
Dr. Berman is former Director of the San Diego Supercomputer Center (2001 to 2009) and former Vice President for Research at Rensselaer Polytechnic Institute (2009-2012). In 2012, she became U.S. lead of the emerging Research Data Alliance (RDA), a community-driven international organization created to accelerate research data sharing world-wide, and to develop the technical, organizational and social infrastructure needed to support data-driven innovation. Dr. Berman is Chair of RDA/U.S. and Co-Chair of RDA’s leadership Council. Prior to joining Rensselaer, Dr. Berman was Professor in the UC San Diego Department of Computer Science and Engineering and first holder of the High Performance Computing Endowed Chair in the Jacobs School of Engineering. As Director of the San Diego Supercomputer Center (SDSC), she led a staff of 250+ interdisciplinary scientists, engineers, and technologists. Dr. Berman directed the National Partnership for Advanced Computational Infrastructure (NPACI), a consortium of 41 research groups, institutions, and university partners with the goal of building national infrastructure to support research and education in science and engineering.
Dorsey Learning Hall 1106, Luddy Hall
Michael Richey | Chief Learning Scientist - The Boeing Company
Advancing Personalized Learning in workplace settings
Abstract: We are living in an age transformed by intelligent systems. Data science including Machine Learning, Artificial Intelligence and Natural Processing Language methods are changing business communication and making transparent workforce expertise and culture. The concept of coupling AI and digital data between agents (learners) the system culture (formal and informal) within a real world environment (robots) enables us to see interaction patterns, agent sensing, acting and knowledge optimizing.
Changes in the skills and knowledge necessary for successful digital life and work flow are now impacting organization cultural systems as well. Organizations are inherently relational. In the jargon of complex systems, firms are complex adaptive systems composed of a network of employees bound together by contracts, who perform tasks by negotiating both internal relationships within firms, but also by observing, managing, and responding to constantly shifting external environments.
These changes driven by convergence methods promise rapid innovation for companies due to the need for a highly-educated “adaptive “workforce, but to also give company’s new opportunities to measure the flow-quality of information including key competency and agility measures. These data are starting to uncover the distributive cognition of agents within a social network, “The ghost in the machine” where thoughts are embodied in digital exhaust of agent actions (Madhavan and Richey, 2016).
Bio: Dr. Michael Richey is the Chief Learning Scientist and a Boeing Associate Technical Fellow currently assigned to support educational technology and innovation research at The Boeing Company. Michael is responsible for leading a team conducting engineering education research projects that focus on improving the learning experience for students, incumbent engineers and technicians. His research encompasses, Sociotechnical Systems, Learning Curves, and Engineering Education Research. The online educational programs and research focus on practical understanding of human learning and the design of technology-enhanced learning environments and promoting global excellence in engineering and learning technology to develop future generations of entrepreneurially-minded engineers. Michael has served on various advisory groups including, the editorial board of the Journal of Engineering Education, Boeing Higher Education Integration Board, American Society for Engineering Education Project Board and the National Science Foundation I-UCRC Industry University Collaborative Research Center Advisory Board. Michael has authored or co-authored over 40 publications in leading journals including Science Magazine, The Journal of Engineering Education and INCOSE addressing topics in large scale system integration, learning sciences and systems engineering. Michael often represents Boeing internationally and domestically as a speaker – presenter and has authored multiple patents on Computer-Aided Design and Computer-Aided Manufacturing, with multiple disclosures focused on system engineering and elegant design. He is currently the Principle Investigator for the Boeing Internet of Learning Consortium, the Technical focal for the Boeing ASEE and manages the Boeing Santa Fee Institute Applied Complexity Network and edX Corporate Advisory Board relationship.
Additional responsibilities include providing business leadership for engineering technical and professional educational programs. This includes development of engineering programs (Certificates and Masters) in advanced aircraft construction, composites structures, systems engineering, product lifecycle management and digital manufacturing. This is achieved by partnering and investing in educational initiatives and programs between industry and institutions of higher learning. Michael has served on various advisory groups including, the editorial board of the Journal of Engineering Education, Boeing Higher Education Integration Board, American Society for Engineering Education Project Board, the National Science Foundation I-UCRC Industry University Collaborative Research Center Advisory Board and has server on the NAE as a committee member for the, “Status, Role, and Needs of Engineering Technology Education in the United States.”. He is currently the Boeing Principle Investigator for the Boeing Internet of Learning Consortium, the Technical focal for the Boeing ASEE and manages the Boeing Santa Fee Institute Action Network relationship. Michael holds a B.A and M.Sc. from ESC Lille in Program Project Management and Ph.D. from SKEMA Business School with a focus on Engineering Education Research. In addition he holds a Caltech Certificate is System Engineering, a Stanford Certified Project Management Certificate and a Masters Certificate in Project Management from Steven’s Institute of Technology.
Dorsey Learning Hall 1106, Luddy Hall
Harish Kunshetty | Data Analytics Leader - Quant Systems Inc.
Can Machine Learning Redefine Customer Experience?
Abstract: Advances in technology and processing power have combined to open
up the possibility of machines that can learn built around artificial
This talk is focused on how the customer experience is being refined with
machine learning through a series of case studies.
Bio: Harish Kunshetty is a Data Management and Advanced Analytics Leader
with 18+ years of experience in Advanced Analytics, Data Governance &
Management, and AI driven Machine Learning technologies.
Harish has worked as a Chief Data Officer at Quant Systems Inc. and
Associate Partner at IBM. Currently, working on data projects in the
Washington DC metro area.
As a North America Cognitive Architect Practice leader at IBM, Harish has
pioneered the development of multiple first-of-kind machine learning
solutions for a number of Fortune 100 clients. He has presented at various
conferences and a featured writer on data management topics.
Dorsey Learning Hall 1106, Luddy Hall
Collective Creativity: How crowds and communities explore design space
Abstract: Online communities and crowds can successfully engage in creative
activity, including the design of products. When they do so, they explore a
space of possible designs. They are affected by when others participate, and
what artifacts they are creating. And they can be affected by the design of
a site, by the recommendations shown to them. By analyzing online
communities, it is possible to gain insight into how and why novelty
appears. Moreover, it is possible to steer the exploration of design space.
Results will be presented from three ongoing NSF-funded projects that
involve both observations of online communities (including Thingiverse,
Scratch, Wikipedia) and experiments with crowds. The end goal of this
research program is to catalyze collective creativity.
Bio: Jeffrey V. Nickerson is professor and associate dean of research in the
School of Business at Stevens Institute of Technology. His research focuses
on different aspects of collective creativity, in particular the way crowds
and communities design digital artifacts: 3D printing designs, systems
designs, source code, and articles. Before joining Stevens, he worked in
industry as a systems designer and software developer: he held positions at
Time Inc, AT&T, Bear Stearns, Salomon Inc., and was a partner at
PricewaterhouseCoopers. He has a Ph.D. in Computer Science and an M.F.A in
Graphic Design. His most recent NSF-funded project looks at the effects of
artificial intelligence on work design: more information can be found at
Dorsey Learning Hall 1106, Luddy Hall
Rahul Basole | Associate Professor and Director - Georgia Institute of Technology
Understanding Complex Ecosystems using Visual Analytics
Abstract:In today’s complex, hyperconnected business environment, successful transformation requires a system perspective. In this talk, I will discuss the challenges in navigating transformations from an ecosystem perspective and present the value of novel data-driven visual analytic and computational methods and tools that can augment and accelerate decision makers’ ability to discover, make sense, and analyze innovation opportunities and competitive strategies. My talk will draw on many examples, including our work on software stacks, APIs, and microservices as well as domain-specific studies such as FinTech and AI. I will present a suite of tools that have been developed in the Enterprise Science Lab, including ecoxight, pulse, graphicle, and epheno. I will discuss challenges associated with the identification and curation of relevant data, novel methods to extract knowledge from unstructured data, and the importance of a human-centered approach to tool development and design.
Bio: Dr. Basole is an Associate Professor in the School of Interactive Computing, the Associate Director for Enterprise Transformation in the Tennenbaum Institute/IPaT at the Georgia Institute of Technology and an affiliated faculty member in the GVU Center and the Health Systems Institute. His research fuses system science and visualization to study technology strategy, innovation management, and transformation of complex enterprise systems. He has received several best paper awards and his work has been published in leading management, engineering, and computer science journals, conference proceedings, and books.
In previous roles, he was the CEO, Founder, and VP Research of a Silicon Valley-based wireless research and consulting firm, the Director of Research and Development at a software firm, and a Senior Analyst at a leading IT management consulting firm. He currently serves as a director or advisor for several technology firms.
He received a B.S. degree in industrial and systems engineering from Virginia Tech, has completed graduate studies in engineering economic systems, operations research, and management information systems at Stanford University and the University of Michigan, and received a Ph.D. degree in industrial and systems engineering from the Georgia Institute of Technology, concentrating in IT and operations management.
Dorsey Learning Hall 1106, Luddy Hall
Nathan Sanders | Chief Scientist
The Role of Inference and Prediction in Data Science for Entertainment
Abstract: The emerging field of data science, and the notion that statistical modeling and machine learning can unlock latent value in data held or acquired by businesses, has already had a transformative impact on many industries and disciplines. In particular, many companies have benefited from accurate predictive systems deployed at scale. Hollywood studios leverage predictive modeling to target advertisements, recommend content, and more. Such predictive applications have proven successful enough that ‘data science’ has in some cases become synonymous with ‘prediction.’ However, inferential applications of modeling are equally important. Statistical inference provides the basis for people to learn from models and improve decision making, in both scientific and industrial contexts. In entertainment, this includes assessing the impact of a blockbuster film’s marketing campaign and inferring the thematic composition of content like TV shows. In this talk, I will introduce applications of both predictive and inferential statistical and machine learning systems in the entertainment industry and highlight the critical importance of investment in interpretation and communication to enable data and models to be leveraged efficiently.
Bio: Nathan Sanders is the Chief Scientist at Warner Media Applied Analytics. He leads a team of social, physical, and computational scientists engaged in deploying data science and human subjects research techniques to better understand and serve consumers of entertainment products including film, TV, and digital video. Prior to that, Nathan built and led the Quantitative Analytics team at Legendary Entertainment’s Applied Analytics division, which was acquired by Warner Media in 2018. Nathan is also a co-founder and Leadership Team Chair of ComSciCon (http://comscicon.org/), the international workshop series on science communication for graduate students. Nathan did his undergraduate work at Michigan State University and earned his PhD in Astronomy and Astrophysics from Harvard University.
Tyler Foxworthy | Chief Scientist - DemandJump Inc.
Analyzing Dynamic Networks with Topological Data Analysis
Abstract: The internet, social networks, the global transportation system… Complex networks are the substrate and substance of modern life. Unlike static networks, the structure and function of these systems evolves dynamically with time, which has traditionally posed significant challenges to analysts seeking to predict their behavior or optimize their performance. In recent years, Topological Data Analysis (TDA) has been demonstrated to be a meaningful platform for computing topological features of a wide variety of complex networks at multiple scales. Although relatively unknown by data scientists working in industry, TDA has over the last decade been gaining strong acceptance in the research community and seen a surge of growth in a wide variety of application areas, particularly when coupled with kernel-based machine learning techniques. The purpose of this talk is to discuss both the theoretical and practical aspects of applying TDA to the study of dynamic networks by demonstrating its application with Python and a years’ worth of real-time airline scheduling data.
Biography: Tyler Foxworthy is a mathematician and computational scientist working on a wide range of problems in machine learning and marketing optimization. An alumnus of Purdue University, his research is largely focused on the development of high performance algorithms for understanding complex networks, machine learning, and natural language processing. Tyler currently serves as the Chief Scientist of DemandJump Inc, and is a scientific advisor to many technology companies and investors in the Midwest. Tyler’s previous experience includes leadership and research roles in academia, biotech, and management consulting. Additionally, Tyler maintains an active research program and is a regular speaker at international scientific and industry conferences.
James "Jimi" Shanahan |
How Gradient and Autodiff are Transforming Deep Learning
Abstract: Just like electricity, the automobile, the Internet, and mobile phones transformed the 20th century, deep learning is transforming the 21st century, changing how people perceive and interact with technology, enabling machines perform a wider range of tasks, in many cases doing a better job than humans. These applications include: voice assistants on our smartphones, product recommendation engines, self-driving cars, deep fakes, high frequency stock market trading, applications for social good (combating crime), playing games (from Go to Atari), preventing credit card fraud, filtering out spam from our email inboxes, detecting and diagnosing medical diseases, the list goes on and on. Large companies, such as Amazon, Apple, Facebook, Google, Microsoft, and venture capitalists alike are investing heavily in deep learning research and applications.
This talk focuses primarily on one of the key enablers of deep learning, that of optimization theory’s gradient descent and its sidekick, autodiff. Shakespeare might have structured such a talk as follows and used the lens of reverse mode autodiff to aid with understanding:
Act 1: Hack it up
Act 2: BackProp: theory to the rescue
Act 3: Layer by layer learning, a medieval pastime
Act 4: Introspection: better init. and activation functions
Act 5: Express-laning the gradient: Skip Connections, the SoTA frontier (LSTMs, ResNet, Highway Nets, DenseNets)
These five acts will be supported by examples and Jupyter notebooks in Python and TensorFlow. In addition, this talk will show how reverse mode autodiff provides an efficient and effective calculus framework that is transforming how we do machine learning and how we should teach it.
Bio: Jimi has spent the past 25 years developing and researching cutting-edge artificial intelligent systems splitting his time between industry and academia. He has (co) founded several companies including: Church and Duncan Group Inc. (2007), a boutique consultancy in large scale AI which he runs in San Francisco; RTBFast (2012), a real-time bidding engine infrastructure play for digital advertising systems; and Document Souls (1999), a document-centric anticipatory information system. In 2012 he went in-house as the SVP of Data Science and Chief Scientist at NativeX, a mobile ad network that got acquired by MobVista in early 2016. In addition, he has held appointments at AT&T (Executive Director of Research), Turn Inc. (founding chief scientist), Xerox Research, Mitsubishi Research, and at Clairvoyance Corp (a spinoff research lab from CMU). He also advises several high-tech startups (including Quixey, Aylien, ChartBoost, DigitalBank you.co, VoxEdu, and others).
Jimi has been affiliated with the University of California at Berkeley and at Santa Cruz since 2008 where he teaches graduate courses on big data analytics, machine learning, deep learning, and stochastic optimization. In addition, he is currently visiting professor of data science at the University of Ghent, Belgium. He has published six books, more than 50 research publications, and over 20 patents in the areas of machine learning and information processing. Jimi received his PhD in engineering mathematics from the University of Bristol, U. K., and holds a Bachelor of Science degree from the University of Limerick, Ireland. He is a EU Marie Curie fellow. In 2011 he was selected as a member of the Silicon Valley 50 (Top 50 Irish Americans in Technology).
that supports HTML5 video
Powered by Events Manager