Recent Activity

2021

Accepted to the 2021 ACM SIGPLAN Machine Programming Symposium (MAPS): "ControlFlag: A Self-Supervised Idiosyncratic Pattern Detection System for Software Control Structures".

Accepted to the 2021 ACM SIGPLAN Machine Programming Symposium (MAPS): "Predictive Locality Optimization for Higher-Order Tensor Computations".

Accepted to 2021 GECCO Workshop on Evolutionary Computation Software Systems (EvoSoft): "AI Programmer: Autonomously Creating Software Programs Using Genetic Algorithms".

44th patent issued: "Extend GPU/CPU coherency to multi-GPU cores" (10,956,330)

43rd patent issued: "Methods and apparatus to detect memory leaks in computing systems" (10,956,298)

42nd patent issued: "Neural network optimization mechanism" (10,929,749)

41st patent issued: "Methods and apparatus for runtime multi-scheduling of software executing on a heterogeneous system" (10,908,884)

Upcoming keynote address @ MIT's DSAIL 2021 virtual retreat: "A Glimpse Into Machine Programming @ Intel Labs"

40th patent issued: "Compute optimization for deep neural networks" (10,902,547)

39th patent issued: "Coordination and increased utilization of graphics processors during inference" (10,891,707)

A video on our ControlFlag system

38th patent issued: "Systems and methods for determining a configuration of a microarchitecture" (10,853,554)

2020

Patent issued: "Systems and methods for determining a configuration of a microarchitecture" (10,853,554)

Invited machine programming talk @ MIT & NeurIPS 2020, ML for Systems Workshop: "A Glimpse Into Machine Programming @ Intel Labs"

Paper accepted to 2020 NeurIPS Computer-Assisted Programming Workshop "Software Language Comprehension using a Program-Derived Semantics Graph"

Paper accepted to NeurIPS 2020 ML for Systems Workshop: "ControlFlag: A Self-supervised Idiosyncratic Pattern Detection System for Software Control Structures"

37th patent issued: "Methods and apparatus to detect anomalies of a monitored system" (10,802,942)

Invited MP talk @ UWisc: "Machine Programming: Challenges and Opportunities"

36th patent issued: "Programmable coarse grained and sparse matrix compute hardware with advanced scheduling" (10,769,748)

Keynote @ Department of Energy's Program Synthesis for Scientific Computing: "Machine Programming: Challenges and Opportunities"

Patent issued: "Neural Network Scheduling Mechanism" (10,719,760)

CACM article on machine programming: "Your Wish is My CMD" by Neil Savage

Paper accepted to MAPL 2020: "Learned Garbage Collection" (joint with MIT)

Paper accepted to CAV 2020: "An Abstraction-Based Framework for Neural Network Verification"

Patent issued: "Detecting Mobile Device Sensor Malfunctions" (10,591,313)

SDTimes & Economic Times highlighting our research.

Patent issued: "Extend GPU/CPU coherency to multi-GPU cores" (10,521,349)


2019

Venturebeat has published an article on my team's machine programming research @ NeurIPS '19!

Patent issued: "Efficient sharing and compression expansion of data across processing systems" (10,497,084)

Interview with Knowledge@Wharton on machine programming.

Open source: AutoPerf (NeurIPS'19) has been released to the open source community.

Patent issued: "Methods and systems for performing a replay execution" (10,474,471)

Intel Newsroom Press Release on my team's Machine Programming Research.

Intel Division Recognition Award: "Outstanding Leadership of Machine Programming Patent Harvest"

Patent issued: "Compute optimization mechanism for deep neural networks" (10,417,734)

Patent issued: "Compute optimization mechanism for deep neural networks" (10,417,731)

Patent issued: "Autonomous machines through cloud, error corrections, and predictions" (10,410,115)

Accepted to NeurIPS: "A Zero-Positive Learning Approach for Diagnosing Software Performance Regressions"

Opening address for Machine Programming Day @ Berkeley: "Intel's Machine Programming Pioneering Research Vision"

Patent issued: "Mechanism for facilitating dynamic and efficient management of instruction atomicity violations in software programs at computing systems"

Patent issued: "Methods and systems to identify and reproduce concurrency violations in multi-threaded programs using expressions"

Accepted invitation as chair of MAPL steering committee.

Q2 Intel Labs' Eureka Award Winner (inventor with most patent applications filed in a quarter (30 new filings)).

Accepted invitation to serve on SysML 2020 program committee.

Patent issued: "Autonomous vehicle advanced sensing and response"

Intel's Annual Gordon Moore Award Nomination: "Informed risk-taking across Intel Labs, PSG, SSG, and University Research that has furthered Intel's FPGA innovations"

      • Category: Excellence in Risk Taking.

      • Team: Aravind Dasu, Mahesh Iyer, Eriko Nurvitadhi, Michael Adler, Justin Gottschlich, Mondira Pant, Todd Younkin

Intel Tech Insights Leadership Award: "Machine Programming: A Radical Approach to Automating Software" (Justin Gottschlich and Tim Mattson)

Patent issued: "Coordination and increased utilization of graphics processors during inference"

Patent issued: "Extend GPU/CPU coherency to multi-GPU cores"

DATSA has been open sourced.

SysML whitepaper: "SysML: The New Frontier of Machine Learning Systems"

Invited talk, Stanford DAWN Retreat '19: "Machine Programming"

Patent issued: "Detecting root causes of use-after-free memory errors"

Patent issued: "Methods and systems to identify and reproduce concurrency violations in multi-threaded programs"

Invited talk to Dawn Song's research team at Berkeley: "Anomaly detection, machine programming, and other AI research at Intel"

Our "Precision and Recall for Time Series" NeurIPS paper has made a few different top paper reading lists. Here's one. Here's another.

Patent issued (milestone, 20th issued patent): "Programmable coarse grained and sparse matrix compute hardware with advanced scheduling."

Co-teaching with Insup Lee and James Weimer: CIS 700-002: Topics in Safe Autonomy, Spring 2019


2018

Invited talk at UW's TVM Conference: "Machine Programming"

Invited talk at Intel's NeurIPS special luncheon: "Anomaly Detection: Today and Beyond"

NeurIPS spotlight talk: "Precision and Recall for Time Series"

NeurIPS 3-minute teaser video: "Precision and Recall for Time Series"

Intel Labs Division Recognition Award for creating and leading the Anomaly Detection IP Think Tank.

Invited talk at SPLASH-I: "The Future of AI: Machine Programmers and Their Necessary Self-Awareness"

Invited talk at Intel's Autonomous Driving Community of Practice Workshop: "Autonomous Vehicles and the Anomalous 1%"

GRASP / PRECISE Industry Symposium at University of Pennsylvania: "Deep Learning for Autonomous Driving" (video here)

MAPL presentation: "The Three Pillars of Machine Programming" (joint with MIT)

Program committee member, SysML 2019.

Invited talk at VMware Research: "Anomaly Detection for Practical Systems (and a Tiny Bit of Machine Programming)"

Special seminar at University of Pennsylvania: "The Future of Anomaly Detection" (slides forthcoming)

General Chair, Second ACM SIGPLAN Workshop on Machine Learning and Programming Languages (MAPL)


2017

Intel's Principal Investigator for the joint Intel/NSF CAPA research center.

Program Chair and Founding Member, First ACM SIGPLAN Workshop on Machine Learning and Programming Languages (MAPL)

Deputy Technical Lead and Founding Member, NSF/Intel Partnership on Computer Assisted Programming for Heterogeneous Architectures


2016

Talk at Intel's High Performance Developers Conference: "Using Machine Learning To Avoid the Unwanted"

Intel Research Velocity Challenge Winner: "Using Deep Neural Networks to Identify and Fix Performance and Correctness Anomalies in Data Centers"

Application Track Chair, TRANSACT 2013