Presented By
O’Reilly + Intel AI
Put AI to Work
April 15-18, 2019
New York, NY

Towards Self-Aware Resilent Systems and Ethical Artifical Intelligence

Pradip Bose (IBM T. J. Watson Research Center), Augusto Vega (IBM T. J. Watson Research Center), Nandhini Chandramoorthy (IBM T. J. Watson Research Center)
4:05pm4:45pm Wednesday, April 17, 2019
Implementing AI
Location: Mercury Rotunda
Secondary topics:  AI case studies, Deep Learning and Machine Learning tools, Reliability and Safety

Who is this presentation for?

Researchers, practioners and executives engaged in defining the future landscape of AI system products and associated software algorithms



Prerequisite knowledge

1) Beginning knowledge about machine learning and deep learning (ML/DL) 2) Beginning (and at most Intermediate) knowledge about software-hardware system architectures

What you'll learn

1) What are the modes and mechanisms of failures in today's AI systems and solutions? 2) How can we make future AI systems more robust and resilient under changing in-field data patterns? 3) What is "Swarm-AI" and how can this technology help us solve the "cognitive resilience" challenges referred to above? What role does the cloud play in this setting? 4) Can human-like qualities of "self-awareness" be imbibed into the cloud-backed swarm-AI paradigm? Can that be used to infuse altruistic behavior in the edge swarm agents so that the swarm computing model enforces standards of ethical decision-making that are approved by human society? The attendees will learn fundamental concepts, along with specific simulation-based implementation demonstrations of the advanced research being pursued in this project.


In this presentation, we will focus on next-generation systems that are proficient in pursuing advanced artificial intelligence (AI) based computation and decision-making. The particular challenge addressed in this ongoing research endeavor is that of cognitive resilience and ethical, bias-free choices. Operating under changing in-field data patterns is difficult in current AI inference engines. Precipitous loss of in-field inferential accuracy can lead to potentially catastrophic errors in decisions. Also, inadequate safeguards in factory-trained model development can lead to decisions at the edge that are perceived as having unintended cognitive bias. How to ensure that the cognitive decision-making at the edge is bias-free, ethical and resilient to in-field data changes or inconsistencies – those are the AI systems challenges we are primarily focused on in this project.

Our research is centered around a visionary system architecture, where the cloud provides continuous, scalable support for a connected swarm of edge compute devices operating in the field of analytics and inferential decision-making at the edge. The particular use cases we are focusing on are: (1) autonomous (self-driving) vehicles; (2) network cyber security through a swarm of intelligent monitoring agents; and (3) distributed, autonomous resource and power management in a data center or server setting. In each of these use cases, we are relying on a novel swarm-AI mechanism to facilitate continuous, in-field learning. Including a simulated “self-awareness” technique within each intelligent edge device, allows interesting new modes of collaborative perception which are not possible with today’s smart robotics technology alone. We present simulation and modeling based results to show the promise of using self-aware swarm-AI technology for advanced cooperative compute-and-infer tasks in a cloud-backed mobile cognition setting.

After establishing the benefit of self-aware swarm-AI technology, we present our ongoing multi-year R&D activity in implementing it for prototype demonstration in a real-life setting for our sponsoring government client.

Photo of Pradip Bose

Pradip Bose

IBM T. J. Watson Research Center

PRADIP BOSE is a Distinguished Research Staff Member and Manager of the Efficient & Resilient Systems Department at IBM T. J. Watson Research Center, Yorktown Heights, NY. He has been involved in the design and pre-silicon modeling of virtually all IBM POWER-series microprocessors, since the pioneering POWER1 (RS/6000) machine, which started as the Cheetah (and subsequently America) superscalar RISC project at IBM Research. From 1992-95, he was on assignment at IBM Austin, where he was the lead performance engineer in a high-end processor development project (POWER3). During 1989-90, Dr. Bose was on a sabbatical assignment as a Visiting Associate Professor at Indian Statistical Institute, India, where he worked on practical applications of knowledge-based (AI)systems. His current research interests are in high performance computers, artificial intelligence, power- and reliability-aware microprocessor architectures, accelerator architectures, pre-silicon modeling and validation. He is the author or co-author of over a hundred publications (including several book chapters) and he also serves as an Adjunct Professor ar Columbia University. He has received twenty five Invention Plateau Awards, several Research Accomplishment and Outstanding Innovation Awards from IBM. Dr. Bose served as the Editor-in-Chief of IEEE Micro from 2003-2006 and as the chair of ACM SIGMICRO from 2011-2017. He is an IEEE Fellow and a member of the IBM Academy of Technology.

Photo of Augusto Vega

Augusto Vega

IBM T. J. Watson Research Center

Augusto Vega is a Research Staff Member at IBM T. J. Watson Research Center. He holds a Ph.D in Computer Architecture from UPC Barcelona in Spain. He is a lead contributor to the swarm-AI project at IBM Research – with specific interests in the self-driving car application space.

Photo of Nandhini Chandramoorthy

Nandhini Chandramoorthy

IBM T. J. Watson Research Center

Nandhini Chandramoorthy is a post-doctoral researcher at IBM T. J. Watson Research Center, Yorktown Heights, NY. She holds a Ph.D degree from Penn State University. Her research interests are in computer architecture, artificial intelligence, power-aware design and modeling methodologies.

Leave a Comment or Question

Help us make this conference the best it can be for you. Have questions you'd like this speaker to address? Suggestions for issues that deserve extra attention? Feedback that you'd like to share with the speaker and other attendees?

Join the conversation here (requires login)