Amazon Research Awards (ARA) provides unrestricted funding and AWS Promotional Credits to academic researchers investigating diverse research topics across multiple disciplines. This cycle, ARA received many outstanding research proposals and today publicly announces 10 awardees representing 10 universities.
This announcement includes awards funded under three calls for proposals during the Winter 2024 and Spring 2024 cycles: AI for Information Security, Foundation Model Development, and Sustainability. The proposals were reviewed for the quality of their scientific content and their potential to influence both the research community and society.
In addition, Amazon encourages the publication of research results, presentations of research at Amazon offices worldwide, and the release of related code under open source licenses.
Recipients have access to more than 300 public Amazon datasets and can use AWS AI/ML services and tools through their AWS Campaign Credits. Recipients are also assigned an Amazon Research Contact, who offers consultation and advice, along with opportunities to attend Amazon events and training sessions.
“Security is critical to Amazon, and artificial intelligence has been instrumental in making progress in this area. The ARA program allows us to engage with the broader academic community to tackle important issues at this intersection of AI and cybersecurity,” said Baris Coskun, Senior Scientist at GuardDuty. “The response to our AI for Cybersecurity call for proposals has been fantastic and we have received a large number of high quality proposals. We look forward to supporting the new recipients in their development of effective new technologies that provide meaningful security value.”
“The response to Amazon’s first Foundation Model CFP was excellent. We awarded the largest Amazon Research Awards grant to date with $250,000 in cloud credits for work on Trainium-enhancing foundational models. Momentum in AI is only getting stronger; with the Build on Trainium program, AWS will invest $110MM to support AI research at universities around the world,” said Emily Webber, Principal Solutions Architect at Annapurna. “We look forward to working with exceptional PIs to develop kernels and algorithms that improve the future of AI for all. The scaling of model growth, in size and applications, provides a strong case for future work at the lowest levels of stack. There’s never been a better time to dive into computer optimization for AI – join us!”
ARA funds proposals throughout the year in a number of research areas. Applicants are encouraged to visit the ARA call for proposals page for more information or send an email to be notified of future open calls.
The tables below show, in alphabetical order by last name, cycle call recipients winter 2024 and spring 2024 sorted by research area.
Spring 2024
AI for information security
| Container | University | Research title |
| Z. Berkay Celik | Purdue University | Time-Preserving Audit Log Reduction: A Scalable Approach to Precise Attack Investigation and Anomaly Detection |
| Kaize Ding | Northwestern University | Label-Efficient Graph Anomaly Detection for Information Security: Detection, Automation, and Explanation |
| Christopher Kruegel | University of California, Santa Barbara | Combating false positives in ML-based security applications with context-aware classification |
| Sijia Liu | Michigan State University | Advancing Reliable Generative AI: The Role of Machine Learning |
| Chongjie Zhang | Washington University in St. Louis | Towards practical preference-based offline reinforcement learning for information security |
| Yue Zhao | University of Southern California | Label-Efficient Graph Anomaly Detection for Information Security: Detection, Automation, and Explanation |
Sustainability
| Container | University | Research title |
| Fengqi you | Cornell University | Large language model co-pilot for transparent and reliable LCA |
Winter 2024
Development of basic model
| Container | University | Research title |
| Lu Cheng | University of Illinois at Chicago | Reliable large-scale model tuning via uncertainty quantification |
| Samet Oymak | University of Michigan, Ann Arbor | Beyond Transformer: Optimal Architectures for Language Model Training and Tuning |
| Hua Wei | Arizona State University | Reliable large-scale model tuning via uncertainty quantification |