Nemertes [Next] – Baltimore 2024

Dragonfly: NASA’s Drone to Titan

Dragonfly may be the most ambitious science mission NASA has ever attempted: sending a car-sized, nuclear-powered octocopter to explore the surface of a distant ocean world. In a voyage straight out of science fiction, Dragonfly will deliver the most expansive suite of science instruments ever dispatched to another celestial body. Dragonfly will explore the organics-rich Titan surface, landing, collecting, and returning results that could change our understanding of life in the universe.

This talk will provide an overview of the mission and discuss some of the challenges involved in planning, designing, fabricating, and testing to assure success of a multi-year, multibillion-dollar NASA mission.

Bio

Dr. Luke A. Becker is Principal Professional Staff at the Johns Hopkins University Applied Physics Laboratory (APL) where he works within the Space Exploration Sector (SES). Luke is a Mission Systems Assurance Manager responsible for leading and managing the mission assurance effort for both Department of Defense and NASA projects. He has worked within the Space Mission Assurance (SMA) group at APL for 23 years. He has led and managed the mission assurance effort on Mini-RF and Parker Solar Probe and was a lead mission assurance team member on NASA's MESSENGER, STEREO, and New Horizons missions. He is currently the Mission Systems Assurance Manager for Dragonfly, NASA's current New Frontiers Mission. Dragonfly will explore Titan to characterize its habitability and investigate the progression of prebiotic chemistry. Dragonfly is slated for a Summer 2028 launch date with arrival at Titan expected late 2034. In addition to his role as a mission systems assurance manager, Luke is also the supervisor of the Spaceflight Systems Assurance Management section within the Space Mission Assurance Group. Luke is also the lead instructor for the capstone, Small Satellite Development and Experimentation, a Space Systems Engineering master's degree at the Johns Hopkins University Whiting School of Engineering and also teaches Facilitation Skills in a master's program at Alabama. Luke is the president of a multi-million-dollar non-profit whose purpose is to provide supplemental funding and support for school initiatives and innovations to benefit the students in the Conewago Valley School District learning community.

Before joining APL, Luke was a quality engineer at the Naval Research Laboratory where he worked on a program for the International Space Station. Prior to that, he served six years active duty in the US Navy, where he worked as a Fire Control technician on shipboard missile systems and their associated radars and power equipment. He holds a B.S. in Quality Systems Management from the National Graduate School, a M.S. in Quality Management from the University of Alabama, and a Doctorate in Organizational Leadership Studies from Northeastern University in Boston, MA.

Telling Stories about Science: How to Do It, Why It Matters

Climate shift, CRISPR, artificial intelligence, cryptocurrencies, quantum decryption, take your pick—the world is full of complexities we must grapple with. What’s more, we as individuals and as societies must make decisions about these complex matters when they are outside of our own expert knowledge. Stories are powerful tools for understanding those complexities and making those decisions. If we want to get those stories right—informed by the real facts, the real risks, the real opportunities—then scientists and other technical experts need to be involved in shaping them. This session will entail a discussion about the why and the how of telling stories about science.

Bio

Erin Bow is a science writer, storyteller, and author with three volumes of poetry and six novels in print – the most recent of which won a Newbery Honor and was longlisted for a National Book Award. She was the storyteller attached to Generation Energy, the Government of Canada's effort to reimagine power generation. She opened for Margaret Atwood's on-stage conversation on "feminism, technology, and the future." She is currently working with the Amelia Earhart estate to re-center Earhart's work as a feminist to a generation that knows her only as a vanished pilot. Her current passion is helping scientists and other highly technical professionals learn to tell better stories.

Bio

John Burke is CTO of Nemertes Research. He conducts and analyzes primary research in, and advises key enterprise and vendor clients on: AI, cloud, networking, infrastructure, and automation, cybersecurity focusing on Zero Trust and SASE/SSE. As a research analyst and consultant, he draws on his 30 years of experience as a practitioner and leader in IT to better understand the needs of IT executives and the challenges facing vendors trying to sell to them. His experience has given him broad and deep knowledge in computing, communications, and IT management.

Mr. Burke holds an MA in the history of science and a bachelor of science in electrical engineering, both from The Johns Hopkins University. His interests include baking, birding, and beer.

Defense in Depth: Safety and Security on the Path to AGI

Major AI labs believe we’re on track to build AGI in the next four years. At the same time, whistleblowers at these labs don’t believe we’re on track to build it safely and securely. If this is true, significant large-scale risks could soon begin to manifest over that timescale, including in geopolitical contexts. This talk will discuss the nature of those risks, possible regulatory and national security strategies to address them, and ways companies and individuals can best prepare themselves. Views are informed by past and present conversations with hundreds of stakeholders from across government and industry, including Congressional offices; Executive Branch leadership and action officers; whistleblowers from frontier AI labs; subject-matter experts across relevant domains; and senior representatives of allied nations. The Chatham House Rule will apply to both the presentation and Q&A segments.

Bio

Edouard Harris is a serial entrepreneur, AI researcher, and national security advisor. Since 2014, Edouard has founded 3 companies, sold one, and raised capital from Silicon Valley's top investors. Edouard holds a PhD in Physics and has 5-10 years' experience variously in software, AI research, and AI safety and national security policy.

Edouard is the co-founder of Gladstone AI, which helps U.S. and allied governments anticipate developments in advanced AI research and their implications for defense and national security. At Gladstone, Edouard led development of a State Department-commissioned Action Plan which outlines a whole-of-government response to national security threats posed by ongoing development of advanced AI. As part of his work at Gladstone, Edouard has briefed over a hundred U.S. government officials including sitting Cabinet Secretaries and Undersecretaries, general officers, Congressional offices, National Security Council staff, and multiple Central Bank Governors.

How to Train Your Robot

Machine learning and artificial intelligence have made rapid advances, but our interaction with them is largely bound to our screens. In this presentation, we will discuss the potential benefits of AI-enabled learning robots, the technical challenges involved, and various modes of effectively teaching robots new skills. Dinesh Jayaraman will use a few projects from his research group's work as case studies to ground this discussion, spanning robot arms, quadrupeds, dextrous hands, and quadrotors.

Bio

Dinesh Jayaraman is an assistant professor at the University of Pennsylvania's CIS department and GRASP lab. He leads the Perception, Action, and Learning (Penn PAL) research group, which works at the intersections of computer vision, robotics, and machine learning. Dinesh received his PhD (2017) from UT Austin, before becoming a postdoctoral scholar at UC Berkeley (2017-19). Dinesh's research has received a Best Paper Award at CORL '22, a Best Paper Runner-Up Award at ICRA '18, a Best Application Paper Award at ACCV '16, the NSF CAREER award '23, an Amazon Research Award '21, and been covered in several press outlets. His webpage is at: https://www.seas.upenn.edu/~dineshj/

The State of Commercial Quantum: Are We There Yet?

Quantum computing, networking, and sensing have gotten a lot of hype over the past few years. But what are enterprise organizations really doing with it? What technology providers are gaining traction (and customers), and why? What universities are generating research that will be usable in the three-to-five-year horizon?

This session provides a detailed review of the state of commercial quantum technology, along with findings from Nemertes’ just-released Emerging Technologies research study (for which we interviewed organizations on their quantum plans).

Bio

Johna Till Johnson is CEO and Founder of Nemertes Research, where she sets research direction and works with strategic clients. Under her leadership, Nemertes has emerged as a leading trusted advisor to Fortune-50 and other world-class organizations.

A widely regarded expert, Ms. Johnson currently serves as the content committee chair at the Wall St. Technology Association (WSTA), which provides technology guidance to some of the world’s largest and most sophisticated financial services firms. She regularly speaks at industry trade shows, conferences, and seminars, including Nemertes Navigator360, IDG’s FutureIT, ISC, and (WSTA). She writes regular columns on cybersecurity and IoT for TechTarget. Prior to founding Nemertes in 2002,

Ms. Johnson served as chief technology officer (CTO) at Greenwich Technology Partners, a global consulting and engineering firm. Prior to that, she led the Global Networking Strategies Service business unit of META Group. Prior to that, as Senior Technology Editor she oversaw the lab-testing program at Data Communications magazine, providing testing and assessment of up to 16 emerging technologies per year. She also has designed and developed security, speech-synthesis, and free-space laser products at companies including Mosler Security Systems and Digitus Corp.

Ms. Johnson holds a bachelor of science degree in electrical engineering/computer science (BSEE/CS) from The Johns Hopkins University, has conducted graduate work in nuclear and particle physics at the University of Rochester, speaks three languages, and has published a science fiction novel. She is an avid urban kayaker and novice woodworker and candymaker.

How to Train Your Robot

Machine learning and artificial intelligence have made rapid advances, but our interaction with them is largely bound to our screens. In this presentation, we will discuss the potential benefits of AI-enabled learning robots, the technical challenges involved, and various modes of effectively teaching robots new skills. Dinesh Jayaraman will use a few projects from his research group's work as case studies to ground this discussion, spanning robot arms, quadrupeds, dextrous hands, and quadrotors.

Bio

Will Liang is a student at the University of Pennsylvania studying computer science. He is conducting research in robotics, reinforcement learning, and computer vision.

Malleable Software in the Age of AI

Today we have thousands of apps to choose from, but it's difficult to craft our own custom tools that do exactly what we need. Geoffrey Litt's research explores malleable software: approaching software that feels more like a Lego set that anyone can combine to create their own tools, without programming. This talk will feature demonstrations of malleable software tools developed in contexts from travel planning to collaborative writing. It will also discuss how AI might enable a Cambrian explosion of custom tools created by non-programmers, and what kinds of new software environments will be needed to take advantage of that new power.

Bio

Geoffrey Litt is a Senior Researcher at the independent research lab Ink & Switch, where he leads the Malleable Software track. His research focuses on developing new tools and media that help people customize software without doing traditional programming. He received a PhD in Computer Science from MIT, working with Daniel Jackson on malleable software.

Empowering Enterprise Efficiency: Generative AI Can Fundamentally Change the Way Teams Work

The promise of enterprise automation has long been a “holy grail” for enterprise leaders, but the cost and complexity of automating workflows has limited the use of automation to only the most critical tasks. Generative AI eliminates these barriers and introduces the possibility to automate repetitive tasks of any complexity. Beyond a simple “digital assistant,” generative AI now has the ability to allow enterprise teams to automate entire workflows, on their own terms, without technical resources or training. Generative AI won't replace enterprise teams; it will make enterprise teams significantly more efficient by delegating tedious tasks to automation.

Bio

Bella Liu is the co-founder and CEO of Orby AI, which provides a generative AI platform to automate repetitive enterprise processes and improve business efficiency. Behind the scenes, the platform is powered by a Large Action Model using a Neuro-symbolic approach combined with an AI agent. The Orby AI team has deep AI and Automation domain expertise coming from Google, UiPath, Microsoft, and Stanford. It is at the series A stage with top investors backing it.

She was previously the AI product leader at UiPath from the early stage to post-IPO. During her tenure, she saw the massive demand for automation and the need for a simpler approach to automating processes. So she teamed up with her co-founder Will Lu, an AI engineering veteran from Google, to rebuild the automation platform with generative AI technologies. Since its release, the product has been used by some of the largest enterprise companies.

Secure Computation during the Classical-to-Quantum Transition

Quantum computing is rapidly advancing, with significant efforts underway across universities, national laboratories, private start-ups, and major corporations. Beyond promising unprecedented computational speedups, quantum technologies have the potential to revolutionize secure communication and computation over both quantum and classical networks.

This talk will explore the transformative applications of quantum technologies in secure computation. While quantum key distribution is a well-known application, numerous other possibilities exist, including certifiable randomness generation, secure multi-party computation, secure cloud quantum computation, and verification of quantum computations. This session will focus on the topic of delegating quantum computations to untrusted quantum cloud servers and ensuring the correctness of their outputs as quantum devices increase in complexity. The discussion will include recent advancements, current limitations, and ongoing challenges in securing computation during the classical–quantum transition, highlighting the path towards a more secure computational future.

Bio

Dr. Mantri is an Assistant Professor in the Computer Science Department at Virginia Tech and a fellow with the Commonwealth Cyber Initiative (CCI). Their research expertise lies in the fields of quantum computation and quantum information, with a particular focus on quantum cryptography. Dr. Mantri works on applying cryptographic methodologies to create secure protocols in the quantum domain and designing classical cryptographic protocols that can withstand the computational power of quantum computers. Their primary research interests include delegated quantum computation and multi-party computation.

Dr. Mantri earned their PhD at the Singapore University of Technology and Design, Singapore. During their postdoctoral years, they worked with prominent researchers at the University of Edinburgh and the Joint Center for Quantum Information and Computer Science (QuICS) at the University of Maryland, College Park. Their work has significantly contributed to advancing the understanding and development of secure quantum and classical cryptographic protocols.

Beyond AI: Where Do We Go From Here?

Today in a world of evolving threats, we rely heavily on artificial intelligence to mitigate cyber weaknesses. This session will explore the intersection of cybersecurity, veracity of data, and generative AI. Cybersecurity is critical to our infrastructure and overall technological architecture, however to be tomorrow-ready, we have to continue to not only combat the conventional threats, but the constant churn of emerging threats uncovered through the use of AI. While AI offers revolutionary capabilities in enhancing threat detection and responsiveness, it also presents new vulnerabilities as our reliance on this game-changing technology increases.

Bio

Don Means Jr. is a Class of 1989 Illinois Institute of Technology graduate. He used his Bachelor of Science in Engineering as the foundation for a stellar career in IT, with 34 years of service, including as an officer in the U.S. Navy and as a civilian in the United States government. He has a Masters Degree in National Resource Strategy from National Defense University and is a Harvard Senior Executive Fellow.

Mr. Means holds a position that is equivalent to a two-star military general and currently serves as a senior executive at the Defense Finance and Accounting Service as the agency’s Chief Information Officer and Director of Information Technology. In this role, he oversees the I&T responsible for enabling the Department of Defense to uphold its fiscal responsibilities for all financial transactions, including payroll and accounting services, enterprise systems for financial transactions, and infrastructure support. Moreover, he ensures that all networks and transactions made through those networks are secure and cyber defended.

Prior to recently joining the Defense Finance and Accounting Service, Mr. Means served as a senior executive and director of the Defense Information Systems Agency’s Operations and Infrastructure Center. Mr. Means oversaw the Defense Information Systems Agency’s largest component, with a global workforce of 6,100 or one-third of the agency’s military, civilian, and contractor personnel. As center director, he led a multi-billion dollar cybersecurity portfolio and, in support of the Department of Defense’s mission, he was responsible for maintaining and defending the world’s third largest IT network next to the United States and China: the Defense Information Systems Network. His center installed and maintained enough optical fiber to wrap around the world more than two times and defended the Department of Defense's global network from hundreds of millions of attacks daily. Without his team's expertise and the capabilities they deliver, the United States cannot fight or support an ally in war.

He has dedicated his professional career to serving the United States and is among few who have the experience of doing so as a member of the armed forces, a civilian and as a member of industry, having spent several years as a senior engineer and analyst with Raytheon.

His commitment to the mission and to driving solutions has been recognized with numerous awards including the Defense Meritorious Service Medal, Black Engineer of the Year Award, and most recently, the Exceptional Civilian Service Award. Mr. Means regularly engages with colleges and universities in support of their efforts to develop the future cyber workforce. To that end, he is currently serving on the Illinois Institute of Technology’s College of Computing Board of Advisors.

The Quantum Cloud: Today’s Landscape and Tomorrow’s Possibilities

This talk explores the current state of quantum networks and their critical role in securing cloud environments. It will address the limitations of quantum cryptography today and highlights integration options within cloud infrastructure. The discussion will cover collaborative efforts with the academy, enterprises, and startups, showcasing solutions that are available today. Lastly, there will be an examination of what the technology makes possible for future implementation. Attendees will gain a comprehensive understanding of how Quantum Key Distribution (QKD) is set to transform cloud security and drive future innovations in the quantum era.

Bio

Juan Moreno is a Principal Technology Evangelist for Quantum Technologies at AWS. Based in Austin, he leads the development and growth of AWS's quantum networking partner ecosystem. With a rich background in infrastructure technical support, Juan has witnessed the technology transformation of both large corporations and nimble startups over the past two decades, from datacenter management to virtualization to cloud computing.

More recently, he has focused on advancing quantum networking, applying his experience to bridge the gap between academia and industry in one of the most challenging and promising areas of the security landscape. A Physics undergraduate student at The Open University, his insights and contributions can be found on AWS blog channels, where he continues to influence and provide practical tips to industry executives and technologists alike.

Adaptive Leadership Training Workshop: Making the Case for Innovation Within the Enterprise

Delivering innovation to an organization is exciting, but is often challenged by budgets, outdated systems, entrenched environments—and last but least, resistance from senior management. Paradoxically, the same individuals who resist a sustained commitment to innovation are the ones pounding the table while insisting that “innovation is the cornerstone of our organization”. In this interactive workshop, Nemertes uses our Adaptive Leadership Training methodology to empower participants to bring innovation to their enterprise and, more importantly, make the case for a sustained investment in innovation to enable ongoing transformation.

Bio

Jerald Murphy is SVP of Research and Consulting for Nemertes. With over three decades of technology experience, Jerry has done everything from neural networking research, integrated circuit design, computer programming, designing global data centers, building core network management for the Internet, to being CEO of a managed services company. Jerry has worked in The United States Army, MCI Communications, META Group, Cognizant, and was the CEO of Banking Infrastructure Technology Services (BITS).

Jerry started his career in the Army, graduating from West Point with a degree in electrical engineering. The Army sent Jerry to Princeton, where he did graduate work in electrical engineering and computer science, specializing in neural network integrated circuit design. He then taught electrical engineering and computer science at West Point and worked at the Army Research Projects Agency (DARPA). On leaving the military, he worked at MCI helping to build the first public Internet backbone, responsible for implementing all the management systems and data centers for MCIs networks. He left MCI to start the network management practice for RPM Consulting. He was then picked to work for META Group, where he ended up running its global Infrastructure, Networking, and Security practice. Jerry was then tapped to work for Cognizant, where he ran its IT consulting practice globally, building some of the world’s largest data centers.

Most recently, Jerry was CEO of ACBB-BITS, a managed services firm that designed, built, and ran computer networks, collaboration, and security systems for financial services companies across the United States. After selling BITS, Jerry is now Senior Vice President of Research and Consulting for Nemertes Research, providing IT strategic consulting for large enterprises and IT service companies.

Renewable Energy On-Demand with Solar Nanomaterials

Climate change mitigation will require strategies beyond converting electricity grids to renewable sources. In 2020, only about 32% of the total U.S. energy-related carbon dioxide emissions came from the electric power sector. Humans use power in a variety of ways, and much of the world’s emissions come from off-grid power use, such as in the transportation sector and from the developing world where large-scale energy grids don’t exist. These forms of power use are only increasing, and, therefore, there is an urgent need for emissions-free technologies that can generate and store energy on-demand and on-the-go, where and when it is needed. This presentation will discuss emerging technologies based on multi-functional nanomaterials that take advantage of the configurable nature and quantum properties of these materials to both harvest and store renewable energy. Specifically of interest are new flexible, lightweight, high-efficiency solar technology platforms with light-collecting capabilities tailored for mobile and other off-grid applications. These technologies, combined with new manufacturing methods, electrode technology, and heat management strategies, should further allow solar power generation and storage functions to be seamlessly integrated into the monolithic systems necessary for powering transportation and mobile systems with the sun.

Bio

Susanna M. Thon is an Associate Professor of Electrical and Computer Engineering and Materials Science and Engineering and the Marshal Salant Faculty Scholar at Johns Hopkins University. Her research is in nanomaterials engineering for optoelectronic devices, with a focus on solar energy conversion and sensing.

Thon's work has received funding from the National Science Foundation, Maryland Energy Innovation Institute, TEDCO Maryland Innovation Initiative, American Chemical Society, and the Department of Defense. She is the recipient of Johns Hopkins' Catalyst and Discovery awards and the National Science Foundation CAREER award, a founding member of the Ralph O'Connor Sustainable Energy Institute (ROSEI) at Johns Hopkins and a past chair of the Optical Society of America's Optics for Energy Technical Group.

Thon received her bachelor's degree from MIT in 2005 and her master's and Ph.D., all in Physics, from the University of California Santa Barbara in 2008 and 2010, respectively. Prior to joining Johns Hopkins in 2013, she worked as a postdoctoral fellow at the University of Toronto.

How Network Effects Have Severed the U.S. Electorate

American politics can seem hopelessly divided and broken, but the reasons why are not always widely understood. Entrepreneur, analyst, and journalist Dave Troy has been studying the American political landscape through the lens of network analysis and has arrived at some original conclusions about what’s gone wrong—and what we might do to regain our footing. His analysis is also rooted in the frames of history, geopolitics, and information warfare, and offers unique insights that transcend party politics to consider the future of dollar hegemony, America’s role in the world, and the conflict between democracy and autocracy globally.

Bio

Dave Troy is an investigative journalist focused on exposing threats to democracy. Based in Baltimore, his background as a technologist with an interest in studying online extremism affords him a unique perspective. His work has appeared at MoMA in New York, and he is a fellow with New America Foundation's Future Frontlines. Dave writes regularly about information warfare, history, and politics. He is the host of the podcast Dave Troy Presents, and speaks regularly at conferences on disinformation, extremism, and information warfare. Contact information is available at davetroy.com.

Adaptive Leadership Training Workshop: Making the Case for Innovation Within the Enterprise

Delivering innovation to an organization is exciting, but is often challenged by budgets, outdated systems, entrenched environments—and last but not least, resistance from senior management. Paradoxically, the same individuals who resist a sustained commitment to innovation are the ones pounding the table while insisting that “innovation is the cornerstone of our organization”. In this interactive workshop, Nemertes uses our Adaptive Leadership Training methodology to empower participants to bring innovation to their enterprise and, more importantly, make the case for a sustained investment in innovation to enable ongoing transformation.

Bio

Donald E. Vandergriff is an award-winning teacher, writer, and lecturer who specializes in military leadership education and training. Vandergriff is the author or editor of seven books and over 100 articles dealing with leadership. His book, Adopting Mission Command: Developing Leaders for a Superior Command Culture has been widely praised and he has twice been recognized as Teacher of the Year for the Army ROTC. Don retired as a Marine and Army officer after 24 years in uniform and another 12 years as a contractor both in the US and overseas. Vandergriff is currently the Director of Adaptive Leadership Training for Nemertes Research.

Fixing Artificial Intelligence via Natural Intelligence

AI today is amazing … until it isn't. In fact, there are two serious limitations with current deployed AI systems. First, they typically do not know how confident they are. So, if you ask them, they can't even respond. If you look under the hood, however, one can determine how certain they are, and it turns out that their uncertainty estimates are way off. Worse, if the query is about something totally unfamiliar to the AI, it is even more certain than for queries that are familiar! In terms of designing an intelligent system, this is the exact opposite of what one would desire. Second, they don't know that time exists, and therefore implicitly assume that past and the future are equivalent. This is problematic, because the future is always (at least somewhat) different from the past, and often relatively easy to predict (e.g., the lion tends to come to the watering hole 5 minutes later than it did the previous day). In this talk, Joshua Vogelstein argues that these two issues—lack of (i) uncertainty quantification, and (ii) time—are actually baked into the foundations of modern AI. In contrast, all of natural intelligence incorporates learning, which does quantify uncertainty and considers time. It does so because uncertainty and time were requirements for survival. He therefore takes a lesson from the natural world and adds these notions into a generalized AI framework. This talk will show how quite simple modifications of existing AI algorithms enable us to address uncertainty quantification and time in problems that are currently intractable for existing AI solutions.

Bio

I am an Associate Professor in the Department of Biomedical Engineering at Johns Hopkins University, with joint appointments in Applied Mathematics and Statistics, Computer Science, Electrical and Computer Engineering, Neuroscience, and Biostatistics. Our research focuses primarily on the intersection of natural and artificial intelligence. We develop and apply high-dimensional nonlinear machine learning methods to biomedical big data science challenges. We have published about 200 papers in prominent scientific and engineering venues, with >12,000 citations and an h-index >45. Our group is one of the few in the world that regularly publishes in both top scientific (e.g., Nature, Science, Cell, PNAS, eLife) and top artificial intelligence (e.g., JMLR, Neurips, ICML) venues. We have received funding from the Transformative Research Award from NIH, the NSF CAREER award, Microsoft Research, and many other government, for-profit and nonprofit organizations.

I have advised over 60 trainees, and taught about 200 students in my eight years as faculty. In addition to my academic work, I co-founded Global Domain Partners, a quantitative hedge fund that was acquired by Mosaic Investment Partners in 2012, and software startup Gigantum, which was acquired by nVidia in early 2022. I live in the Chesapeake Bay Watershed with my beloved eternal wife and our three children.

How to Train Your Robot

Machine learning and artificial intelligence have made rapid advances, but our interaction with them is largely bound to our screens. In this presentation, we will discuss the potential benefits of AI-enabled learning robots, the technical challenges involved, and various modes of effectively teaching robots new skills. Dinesh Jayaraman will use a few projects from his research group's work as case studies to ground this discussion, spanning robot arms, quadrupeds, dextrous hands, and quadrotors.

Bio

Sam Wang is a student at the University of Pennsylvania, majoring in Artificial Intelligence. He is a member of the Perception, Action, and Learning group where he has conducted research on how foundation models, such as large language models (LLMs), can be used to train robots. Previously, Sam was awarded the 2nd Grand Prize at Regeneron ISEF 2023 for developing a deep learning algorithm for rapid cardiac screening.

Context is Next

Steven Hawking said that the enemy of knowledge isn’t ignorance, it’s the illusion of knowledge. And we all know where AI’s failings fall in that analysis. For AI to fulfill its potential, its answers mustn’t just sound sensible, they must be sensible. Context is how we make sense as humans and will have to be how we make sense in AI.

Bio

K Krasnow Waterman is a polymath, with a history of weaving broad knowledge to rapidly create solutions or transform data-driven entities. She spent years as a "fire jumper" before becoming a Board Director and business mentor. Then, she was known for jumping into the most challenging situations, quickly building or revamping organizations in data-intensive environments, introducing relevant innovation while reducing cost and risk.

She served as a COO, CIO, CTO, Counsel in large scale organizations like Citigroup, the FBI, and a Ranieri private equity firm, as well as founded, managed, and advised small and mid-sized businesses. Direct budgets from $100k to $1B and teams of 2 to 1,500, matrix-management of 10,000+.

Now, she is sought after to share that breadth of knowledge and experience to help others who are building or scaling ventures, ranging from Board Director & Risk Committee Chair for Cross River ($8B fintech) to Entrepreneur in Residence for a multi-disciplinary university incubator to one-time mentor to hundreds of startups. She has helped businesses in verticals that might be expected - FinTech, Crypto, AI & other data analytics, Cybersecurity, Privacy, LegalTech - and the unexpected - pico satellites, natural resource discovery, carbon accounting, interactive chairs, shopping avatars and more. K is also known for her fast-paced, data-packed, and entertaining talks.

At MIT, K does policy reasoning, privacy, security, and other AI research.  She’s also recently written a new piece of genAI software to dynamically generate context.