Erik Hemberg

Over 15 years of award-winning expertise in the research, development and application of Artificial Intelligence, Machine Learning and Data Science to improve performance in the fields of cybersecurity, program synthesis, tax compliance, learning, and neuroscience.

  • Published > 150 peer-reviewed papers with > 1, 800 citations and > 20 international conference presentations, multiple paper awards.
  • Organized > 10 international workshops and conferences.
  • Reviewed > 300 papers for > 30 different journals and conferences. Editorial board member on 3 academic journals.
  • Invited to speak about cutting edge AI research at > 30 universities, conferences and companies.

Biography

Research scientist in the ALFA group at MIT CSAIL. PhD in Computer Science from University College Dublin. MSc in Industrial Engineering and Management from Chalmers University of Technology, Sweden. Postdoc at MIT.

  • Secured competitive research funding through grant writing. Projects with OpenAI, US Government Agencies, MIT-IBM AI Alliance, DARPA MIPS, COMBAT, XD3, CHASE, IDAS programs, MIT Lincoln Labs , MIT-HKUST Research Program, USAF-MIT AIA, FinTech@CSAIL and CyberSecurity@CSAIL.
  • Created BRON, a knowledge graph linking cyber attacker behavior, vulnerabilities, exploits and mitigations for efficient querying, reasoning and knowledge discovery. It is also used to fine tune Large Language Models on Cybersecurity tasks
  • Improved the robustness and reduced computational complexity of Generative Adversarial Network (GAN) and Autoencoder (AE) training with spatial coevolutionary training
  • Mentored graduate (35) and undergraduate students (49).
  • Facilitate bi-annual executive training for 30 members of Bank of America’s CISO leadership team on the topic of AI Applied to Cybersecurity
  • Used grammar based Genetic Programming with novelty search to improve program synthesis performance on benchmarks compared to the existing State-of-the-Art by incorporating more domain knowledge and exploration
  • Developed and delivered a Blended Online course in Evolutionary for Shantou University in China.
  • Led innovative research for AI and tax administration in the Simulating Tax Evasion and Law with Heuristics project with MITRE Corp, Computer Scientists Wield Artificial Intelligence to Battle Tax Evasion in the New York Times, October 9, 2015.

Publications

Google Scholar Profile
ORCID Profile: 0000-0001-8905-1186