Brakeing Down Security Podcast   /     Machine_Learning- Market Hype, or infosec's blue teams newest weapon?

Description

Direct Link: http://traffic.libsyn.com/brakeingsecurity/2017-026-Ally_miller_machine-learning-AI.mp3 Ally Miller (@selenakyle) joined us this week to discuss Machine Learning and #Artificial #Intelligence. It seems like every new security product employs one or both of these terms. She did the keynote at Bsides Las Vegas on topics of #Machine #Learning and #Behavioral #Economics. We asked Ms. Miller to join us here to discuss what ML and AI are, how algorithms work to analyze the data to come to the right conclusion. What is required to get a useful algorithm, and how much or little human interaction is required? We also discuss a bit of history with her, how IDS/IPS were just dumber versions of machine learning, with 'tweaks' being new Yara or snort rules to tell the machine what to allow/disallow.  Finally, we discussed how people who are doing our 2017 DerbyCon CTF, instructions on how to win are in the show, so please take a listen.   RSS: http://www.brakeingsecurity.com/rss Youtube Channel:  https://www.youtube.com/channel/UCZFjAqFb4A60M1TMa0t1KXw #iTunes Store Link:  https://itunes.apple.com/us/podcast/brakeing-down-security-podcast/id799131292?mt=2  #Google Play Store: https://play.google.com/music/m/Ifp5boyverbo4yywxnbydtzljcy?t=Brakeing_Down_Security_podcast     Join our #Slack Channel! Sign up at https://brakesec.signup.team #iHeartRadio App:  https://www.iheart.com/show/263-Brakeing-Down-Securi/ #SoundCloud: https://www.soundcloud.com/bryan-brake Comments, Questions, Feedback: bds.podcast@gmail.com Support Brakeing Down Security Podcast on #Patreon: https://www.patreon.com/bds_podcast #Twitter: @brakesec @boettcherpwned @bryanbrake @infosystir #Player.FM : https://player.fm/series/brakeing-down-security-podcast #Stitcher Network: http://www.stitcher.com/s?fid=80546&refid=stpr #TuneIn Radio App: http://tunein.com/radio/Brakeing-Down-Security-Podcast-p801582/           show notes   what is the required amount of data required to properly train the algorithms   how do you ensure that the training data is clean (or perhaps how do you determine what causes a false positive or negative)   Xoke Soru: "why are you trying to make skynet and kill us all?  Do you hate humanity?"   Who will ML replace? Who in security?   Ask why people get confused between AI and Machine learning, and where the fine line is between the two or is one actually a subset of the other.   Basically.. "in what way/how do you see ML being used in an offensive capacity in the future (or now)"   https://en.wikipedia.org/wiki/Artificial_neural_network   https://en.wikipedia.org/wiki/Machine_learning   https://en.wikipedia.org/wiki/Portal:Machine_learning   https://www.slideshare.net/allyslideshare/something-wicked-78511887   https://www.slideshare.net/allyslideshare/201209-a-million-mousetraps-using-big-data-and-little-loops-to-build-better-defenses   https://conferences.oreilly.com/velocity/vl-ca/public/schedule/detail/61751   O’Reilly Conference 31 October   Mick douglas class Derbycon CTF Book club   Patreon slack

Summary

Ally Miller (@selenakyle) joined us this week to discuss Machine Learning and #Artificial #Intelligence. It seems like every new security product employs one or both of these terms. We asked Ms. Miller to join us here to discuss what ML and AI are, how algorithms work to analyze the data to come to the right conclusion. What is required to get a useful algorithm, and how much or little human interaction is required?