Artificial

Intelligence

Research Agenda

The goal of J.P. Morgan AI Research is to explore and advance cutting-edge research in the fields of AI and Machine Learning, as well as related fields like Cryptography, to develop solutions that are most impactful to J.P. Morgan’s clients and businesses.

The team is headquartered in New York and present in key hubs around the world.

Data & Cryptography

New ways to clean, integrate and auto-generate appropriate data to help train our machine learning algorithms, while committing to meet the highest security and privacy standards

Learning From Experience

Through techniques such as deep learning and reinforcement learning, not only can machines learn, but also humans can learn what the machines are capable of

Explainability & Interpretability

Our clients, communities and regulators must be able to understand our algorithms. Our team collaborates closely with risk and technology to ensure interpretable techniques

Ethics & Fairness

Our researchers are focused on creating unbiased and ethical models in order to sustain the firm’s relationship with its clients, communities and regulators

Leadership

J.P. Morgan AI Research has assembled a team of experts in various fields of AI. They pursue primary research in areas relative to our research pillars as well as concrete problems related to financial services. They partner with various internal teams to accelerate the adoption of AI within the firm. They also work with leading faculty around the world on areas of mutual interest.

Manuela Veloso, PhD

Head of AI Research, J.P. Morgan

Manuela M. Veloso is the Head of J.P. Morgan AI Research, which pursues fundamental research in areas of core relevance to financial services, including data mining and cryptography, machine learning, explainability, and human-AI interaction. J.P. Morgan AI Research partners with applied data analytics teams across the firm as well as with leading academic institutions globally.

Professor Veloso is on leave from Carnegie Mellon University as the Herbert A. Simon University Professor in the School of Computer Science, and the past Head of the Machine Learning Department.

With her students, she had led research in AI, with a focus on robotics and machine learning, having concretely researched and developed a variety of autonomous robots, including teams of soccer robots, and mobile service robots. Her robot soccer teams have been RoboCup world champions several times, and the CoBot mobile robots have autonomously navigated for more than 1,000km in university buildings.

Professor Veloso is the Past President of AAAI, and the co-founder, Trustee, and Past President of RoboCup. Professor Veloso has been recognized with a multiple honors, including being a Fellow of the AAAI, AAAS, ACM, IEEE. She is the recipient of several best paper awards, the Einstein Chair of the Chinese Academy of Science, the ACM/SIGART Autonomous Agents Research Award, an NSF Career Award, and the Allen Newell Medal for Excellence in Research.

Manuela Veloso's Publications

Sameena Shah

AI Research Director

Full Bio

Prashant Reddy

AI Research Director

Full Bio

Tucker Balch

AI Research Director

Full Bio

Prashant Reddy

AI Research Director

Prashant Reddy is a Managing Director for AI Research at J.P. Morgan in New York, where he is a Research Director and the Head of Technology for AI Research. Previously, Prashant was at Google where he led the Machine Learning team for Android/IoT. Prior to that, he was Senior Algorithms Manager at Nest Labs and the Founder & CEO of Lumator, a tech startup that provided AI-powered services to consumers in electricity markets. At Carnegie Mellon University, he co-created Power TAC, an open-source agent-based simulation environment for Smart Grid research and competition. Earlier in his career, Prashant was a Managing Director at Morgan Stanley where he led engineering teams for algorithmic trading and distributed trading and risk infrastructure. Prashant is a winner of the McGinnis Venture Competition, the Canfield-Roseman Entrepreneur of the Year Award, and the Contributions to the Morgan Stanley Franchise Award.

Prashant has received PhD and MS degrees from the Machine Learning Department at Carnegie Mellon University, an MBA in Finance & Management from the Wharton School at the University of Pennsylvania, and a BS in Electrical Engineering & Computer Science from the University of California at Berkeley.

Tucker Balch

AI Research Director

Dr. Balch is a Research Director at J.P. Morgan AI Research and a professor of Interactive Computing at Georgia Tech (on leave). He is interested in problems concerning multi-agent social behavior in domains ranging from financial markets to tracking and modeling the behavior of ants, honeybees and monkeys. He co-founded Lucena Research, an investment software firm that applies Machine Learning and Big Data approaches to investment problems. Balch has published 120 peer-reviewed articles.

His work has been covered by the Wall Street Journal, CNN, New Scientist, Institutional Investor, and the New York Times. His graduated students work at NASA/JPL, Boston Dynamics, Goldman Sachs, Morgan Stanley, Citadel, AQR, and BlackRock. Before his career in computing, Tucker was an F-15 pilot in the US Air Force.

Sameena Shah

AI Research Director

Sameena Shah is a Managing Director in the AI Research organization at J.P. Morgan. She is a highly accomplished technology leader with over 20 years of educational and industry experience in engineering, AI, and leading development teams that created top AI technologies in the world for financial, news, commodities and legal businesses.

Previously, Sameena was Managing Director, Head of Data Science at S&P Global Ratings where she led the firm’s strategy and development for Augmented Intelligence. Prior to that, Sameena worked at Thomson Reuters for seven years in roles of increasing responsibility that involved building state of the art AI systems that resulted in business growth and operational efficiencies. Sameena is also the Founder and CEO of Aylan Analytics LLC, and has worked at Yahoo! Research, a NYC based hedge fund, an International hedge fund, and a global startup.

Sameena has a PhD in Distributed Machine Learning and a Masters in Computer Science from IIT Delhi. She is the winner of the top PhD in the country award, Cloudera top AI/ML application award, several best paper awards and recognitions. She has contributed 41 Publications, and 11 Patents.

Life as an AI Researcher & Machine Learning Engineer

Our J.P. Morgan office is a place where you can solve real-world problems using state of the art machine learning methods and cutting-edge AI research.

Created with Sketch.

Manuela Veloso: So, at J.P. Morgan, the interesting thing is that we are a firm that has been around for a long time. But it's a firm that has a lot of appetite.

Ashleigh Thompson: One thing's for sure, no two days here ever look the same. I like to start my day in London early. Since we're a global team, it gives me the chance to review work our New York team did last night and catch up live with my colleagues in India.

Virgile Mison: The Machine Learning Center of Excellence develops and deploys machine learning models across different trading and IT platforms of J.P. Morgan.

Saket Sharma: J.P. Morgan, as a bank, has been incorporating machine learning into a lot of our work flows. So, as a Machine Learning Engineer, this is a great time to work on problems with firmwide impact.

Samik Chandarana: We need humans and AI to work together because ultimately, having and learning from what people are doing today in the processes they do and how they operate today provides a great amount of information of how we design systems of the future.

Andy Alexander: External conferences are really important for a number of reasons. One - it allows us to bring in the best of academia and external thought to the organization. The other is that it allows the team to go out to continue to learn. We rely a lot on where we're going, as well as where we've been.

Lidia Mangu: So, we come back from a conference knowing where the field is. And how, you know, taking those state of the art methods and applying them to the problems in the bank.

Simran Lamba: The most exciting and novel thing about working with AI Research is getting to publish our work at the most esteemed academic conferences like ICML, AAAI, and NeurIPS. We not only participate, but we also host and sponsor workshops at these conferences.

Naftali Cohen: I get to focus on the hot topics in AI and machine learning, such as reinforcement learning, cryptography and explainability.

Ashleigh Thompson: Millions of people use and rely upon our products and services every day. Working here, you have the ability to be on the forefront of changing that interaction.

Manuela Veloso: We apply and discover new AI techniques to handle complex problems such as trading, multi-agent market simulations, fraud detection, anti-money laundering and issues related to data.

Virgile Mison: As a technologist I was the most surprised by the wide variety of problems that we have to tackle and that J.P. Morgan is in the unique position to solve thanks to the large amount of data available.

Samuel Assefa: We focus on a number of research problems. One of the most exciting ones is ensuring that AI models are explainable, fair and unbiased.

Andy Alexander: In my life span, I don’t expect to see generalized AI become something that's mainstream. And so for a lot of time we're expecting to see humans and machine helping each other.

Lidia Mangu: Every day is different. Every day we get a new challenging problem. Sometimes there is no known solution for that problem and it is like a new puzzle. Sometimes there is a known solution, but we show how we can do better using state of the art machine learning techniques.

Manuela Veloso: There is a lot of belief as we move that AI and machine learning is this one-shot deal. We do it, we are done. We'll never be done.

Naftali Cohen: I work with some of the best and most creative minds in the field and I have ownership over my work which is very rewarding.

Naftali Cohen: I'm researching how to apply innovative computer vision and deep learning techniques to understand the complexity of decision making in the financial market and recommend clients for market opportunities.

Simran Lamba: What excites me the most about my job here, in New York, is the opportunity to learn from our leaders and external professors. And my favorite part of the day would be brain-storming creative research ideas to solve challenges across all lines of businesses.

Simran Lamba: I’m currently using event logs of Chase customers called ‘Customer journeys’ to find ways to create an even better experience for our clients.

Manuela Veloso: We do believe that junior people are the ones, in some sense, that have that vision. That can think big and that they are not kind of like constrained.

Samik Chandarana: Our clients are getting younger they want to be interacting in different ways and we need fresh talent to come up and help us with those new ideas and actually implement them in a way that makes sense for the client experience.

Lidia Mangu: The advice I would give to a junior executive is to be open-minded. Not to be afraid to learn new things every day. The field is moving very fast.

Virgile Mison: There are many opportunities to learn at J.P. Morgan. Like collaborating with experts in natural language processing, deep learning, time series and reinforcement learning.

Ashleigh Thompson: I'm excited to be part of the transformation to a truly data-driven culture.

END

Engagements with Academia

The firm has launched the first ever J.P. Morgan AI Research Awards Program this year. This effort will continue to foster current partnerships and build new relationships with the best-in-class universities, students, and researchers.

J.P. Morgan Distinguished Lecture Series on AI

J.P. Morgan gathers global AI experts to share expertise in an ongoing series.

Press

Sponsored Events

Learn more about the upcoming conferences we’re attending and sponsoring

December 2019: Vancouver, Canada


Robust AI in Financial Services: Data, Fairness, Explainability, Trustworthiness and Privacy

J.P. Morgan served as a platinum sponsor for the Thirty-third Conference on Neural Information Processing Systems (NeurIPS) 2019 conference held in Vancouver, Canada. NeurIPS has established itself as one of the leading AI conferences in the world, with a record 13,000 attendees this year. To mark its growing interest in technology investment, AI and machine learning, J.P. Morgan sent nearly 50 employees to the conference with representation from all lines of business and three continents.

This year, the firm also hosted its first industry-focused EXPO workshop in partnership with Two Sigma and Hudson River Trading, which garnered over 450 attendees, filling every available seat in the room. In total, 14 research papers authored by JPMC employees were accepted to the workshop. With the growing interest and strong attendance in finance workshops, Tucker Balch and Manuela Veloso from the AI Research team are helping to organize the first international AI in Finance (ICAIF) conference in partnership with the Association for Computing Machinery (ACM), which will be held in New York City on October 15-16 in 2020.

December 2018: Montreal, Canada

Challenges and Opportunities for AI in Financial Services: the Impact of Fairness, Explainability, Accuracy, and Privacy

With nearly 9,000 attendees, NeurIPS is one of the most prominent and longest-running machine learning and artificial intelligence conferences in the world, bringing together academics and tech giants like Google to share cutting-edge research.

For the first time since its inception 32 years ago, the conference hosted a workshop on AI in financial services. This workshop was jointly organized by J.P. Morgan and Capital One, with invited speakers from Georgia Tech, University of Michigan, Cornell, S&P Global.

J.P. Morgan’s head of AI Research Manuela Veloso and head of CIB Data Analytics, Applied AI & ML Samik Chandarana spoke at the workshop, which addressed key challenges to the adoption of AI in the financial system:

  • Ethical and regulatory needs to prove that models are fair and unbiased
  • Requirements around explainability and interpretability of models
  • Operating standards which require nothing short of perfect accuracy for certain use cases
  • Privacy issues around collection and use of consumer and proprietary data

In addition to the workshop, Veloso and Chandarana, joined by other J.P. Morgan employees, held “lightning talks” at the J.P. Morgan booth attracting hundreds of candidates and hosted a reception attracting leading AI researchers from industry and academia.

June 2019: Long Beach, California

J.P. Morgan sponsored the 2019 International Conference on Machine Learning (ICML) in Long Beach, California. ICML is a premier gathering of over 6,000 participants dedicated to the advancement of the branch of artificial intelligence known as machine learning. Participants span a range of backgrounds, from academic and industrial researchers, engineers and entrepreneurs, to graduate students and PhD postdocs. Professionals present and publish cutting-edge research on all aspects of machine learning.

This year, J.P. Morgan ramped up its presence by serving as a gold sponsor, sending 30 employees from across the firm on different teams to share and discuss the firm’s commitment to this space. This is a reflection of the firm’s growing commitment to key investments in AI and machine learning and it’s starting to pay off. Attendees were eager to hear how the firm is applying AI within the firm and what opportunities J.P. Morgan has available.

The firm hosted a workshop on AI in Finance, the first of its kind in the conference’s 36 year history. J.P. Morgan employees presented six research papers and garnered over 300 workshop attendees.

July 2019: Sydney, Australia

J.P. Morgan was a proud sponsor of RoboCup 2019.

June 2018: Montreal, Canada

4,000 participants from 39 countries traveled to Montreal to compete in one of the most anticipated robotics and machine learning competitions of the year - RoboCup 2018.

RoboCup, co-founded in 1996 by Manuela Veloso tasks participants with building and training robots to compete against one another in a real- life, autonomous soccer tournament. Over the past 22 years, extreme advancements have been made to not only the robots, but also the fields in which they compete in, giving the competitors more complex obstacles to overcome. The competition's ambitious goal is to inspire the development of robots that can compete against humans in the actual soccer World Cup by the year 2050.

The week was split into two competitions: the junior tournaments for those ages 12-19, and major tournaments for those over the age of 19. Although most of the 4,000 participants comprise of top talent in the machine learning, artificial intelligence and robotics space, students don't need a degree in STEM to enjoy this competition. In addition to soccer, during the competition robots compete to rescue, work around homes, and even have dance competitions in addition to the soccer matches.

During the rescue competition, robots were tasked with rescuing 'humans,' where they had to successfully find and identify the parts of the obstacle course that stimulated human body temperature. In the future, these kinds of robots could aid first responders in times of crisis- whether that be during a fire, earthquake, or bombing. The home robots were tasked with more butler-type of activities, like grabbing an orange juice off the table, or opening the dishwasher or walking into the bedroom.

J.P. Morgan's global sponsorship of RoboCup is a testimony to the firm's dedication to technology, machine learning and applied data science efforts and AI Research.

May 2019: Montreal, Canada

J.P. Morgan was a proud sponsor of AAMAS 2019.

July 2019: Baltimore, Maryland

Check back after the event for a full recap.

August 2019: Santa Barbara, California

J.P. Morgan is a proud sponsor of CRYPTO. Join us for our workshop, “Privacy Preserving Machine Learning.” Check back after the event for a full recap.

Related Insights

 

Copyright © 2020 JPMorgan Chase & Co. All rights reserved.