Detecting Anomalies with ‘Project Caspian’

Podcast: Play in new window | Download

Subscribe: Apple Podcasts | RSS

TechSpective Podcast Episode 126

Between the persistent cadence of new technology and expansion of the attack surface, and the constant evolution of the threat landscape, organizations face a daunting task. Cybersecurity teams are faced with an overwhelming amount of information and alerts and it’s their job to find the proverbial “needle in a haystack.” Except, the needle isn’t in a haystack. The needle is in a large pile of other needles–and the objective is to figure out which needles matter or which needles post the most significant threat so they can be prioritized and addressed.

No problem, right?

Well, no. It is very much a problem. Thankfully, we have machine learning (ML) to do the heavy lifting. ML can be used to sift through mountains of data in a fraction of a fraction of the time it would take a human (or humans) to do so manually, and quickly identify events that deserve greater scrutiny. It gives cybersecurity teams a manageable starting point.

That is the goal of Adobe’s “Project Caspian.” There are two research papers related to “Project Caspian” if you want to get down into the weeds and understand the details behind it. I perused the research papers, but I decided it would be better and easier to just invite Hayden Beadles, Senior Security ML Engineer, and Jericho Cain, Senior Staff Data Scientist, to join me on the TechSpective Podcast to talk about “Project Caspian.”

Check out the full episode for more on detecting anomalies and improving cybersecurity with machine learning. There is also a brief digression when I suggest that we need a whole separate podcast with Jericho just to chat about physics and a plug for Andy Weir books (“Project Hail Mary” is the title I couldn’t think of during the podcast. Excellent book. I highly recommend it).

The podcast itself is audio-only, but the video of our conversation is also available on YouTube if you prefer:

Please ask questions and share your thoughts on the topic in the comments below. Also, please subscribe to the TechSpective Podcast through your favorite podcast platform and share the podcast with your peers and friends.

If you enjoy the podcast, I would also be grateful if you could take 2 minutes to rate and review the podcast on iTunes or wherever you listen.