It was great to talk about reproducible workflows for @riotscienceclub @riotscience_wlv. You can watch the recording below, but if you don't want to listen to me talk for 40 minutes, I thought I would summarise my talk in a thread:

A Thread about Famous Armenian Inventors and what we have contributed to the world. So the next time someone wants to say we are unimportant, you can rest easy knowing Armenians have contributed some of the greatest inventions to mankind. Here is a handful but there's more.

Raymond Damadian: Armenian-American Physician and inventor of the first MRI machine. MRI is a medical imaging technique used in radiology to form pictures of the anatomy and the physiological processes of the body. Raymond created the first MRI in 1977.

Artem Mikoyan: One of the most famous engineers in aviation. Mikoyan along with Gurevich created one of the most famous Jet lines in history, the MiG. MiG jets directly competed with American Lockheed Martin F Series planes and were among the first to surpass the speed of sound.

Hovannes Adamian: Most notably laid the foundations of color television with his Tri-Color Principle. His work would go on to become the framework that is used to create Color-TV that we see today. He is recognized as one of the founders of color television.

Emik Avakian: An inventor who unfortunately was disabled, but that didn't stop him from contributing work that would help people similar to his condition. Notable contributions are the automatic wheelchair which dismiss the action of hand rolling and a breath operated typewriter.

Next up at #enigma2021, Sanghyun Hong will be speaking about "A SOUND MIND IN A VULNERABLE BODY: PRACTICAL HARDWARE ATTACKS ON DEEP LEARNING"

(Hint: speaker is on the

In recent years ML models have worked from research labs to production, which makes ML security important. Adversarial ML research studies how to mess with ML

For example by messing with the training data (c.f. Tay which became super-racist super-fast) or by foiling ML models by changing inputs in ways humans can't see.

Prior work considers ML models in a standalone, mathematical way
* looks at the robustness in an isolated manner
* doesn't look at the whole ecosystem and how the model is used -- ML models are running in real hardware with real software which has real vulns!

This talk focuses on hardware-level vulnerabilities. This is particularly interesting because these can break cryptographic guarantees (because those are outside of their threat models)
e.g. fault injection attacks, side-channel attacks