Bobby Filar

Hello, I’m Bobby Filar, a machine learning researcher and currently the Head of Data Science at Sublime Security, the first truly open email security platform. I lead our efforts in integrating machine learning capabilities into our product, aiming to make it more intuitive for users and combat sophisticated attacks. Before joining Sublime Security, I was at the helm of Security Machine Learning teams at Endgame and Elastic. During my time at Endgame, I introduced Artemis, a groundbreaking intelligent assistant for cyber defense. This innovative tool allowed security analysts to interact with endpoint data through natural language, streamlining the investigation process and significantly enhancing decision-making in cybersecurity. My work has consistently focused on SecML research, particularly in malware classification and assistive ML.

My passion lies in augmenting security workers’ day-to-day operations through:

Natural language understanding
Reinforcement learning
Adversarial machine learning

I’m driven by the challenge of developing solutions that not only safeguard data but also empower security professionals to work more efficiently and effectively.

selected publications

The malicious use of artificial intelligence: Forecasting, prevention, and mitigation

Miles Brundage, Shahar Avin, Jack Clark, and 8 more authors

arXiv preprint arXiv:1802.07228, 2018

Bib

@article{brundage2018malicious,
  title = {The malicious use of artificial intelligence: Forecasting, prevention, and mitigation},
  author = {Brundage, Miles and Avin, Shahar and Clark, Jack and Toner, Helen and Eckersley, Peter and Garfinkel, Ben and Dafoe, Allan and Scharre, Paul and Zeitzoff, Thomas and Filar, Bobby and others},
  journal = {arXiv preprint arXiv:1802.07228},
  year = {2018},
  url = {https://arxiv.org/abs/1802.07228},
  abs = {https://arxiv.org/abs/1802.07228},
}

Learning to evade static pe machine learning malware models via reinforcement learning

Hyrum S Anderson, Anant Kharkar, Bobby Filar, and 2 more authors

arXiv preprint arXiv:1801.08917, 2018

Bib

@article{anderson2018learning,
  title = {Learning to evade static pe machine learning malware models via reinforcement learning},
  author = {Anderson, Hyrum S and Kharkar, Anant and Filar, Bobby and Evans, David and Roth, Phil},
  journal = {arXiv preprint arXiv:1801.08917},
  year = {2018},
  url = {https://arxiv.org/abs/1801.08917},
  abs = {https://arxiv.org/abs/1801.08917},
}

Ask Me Anything: A Conversational Interface to Augment Information Security Workers

In Thirteenth Symposium on Usable Privacy and Security (SOUPS 2017), Jul 2017