Coursera
SubscriptionAI Safety – Full Course by Safe.AI Founder on Machine Learning & Ethics (Center for AI Safety)
No ratings yetBEGINNEREnglish
YouTubeYouTubePrice
This course is free to watch on its platform.
Watch free on YouTube →Mark your own learning status. Opening a course link does not establish enrollment.
Loading saved status…
The description organizes the course around robustness, which is withstanding hazards; monitoring, which is identifying them; alignment, which reduces hazards inside ML systems; and systemic safety. Existential risk from strong AI is the frame that ties these together.
Chapters run from risk decomposition and accident models through adversarial robustness, anomaly detection, transparency and trojans. Later sections cover detecting emergent behavior, honest models, machine ethics, ML for cyberdefense, cooperative AI and an X-risk overview.
The chapter list runs past the eight-hour mark, so this is a long course to work through in stages. Chapter markers make it easy to jump straight to one topic, such as adversarial robustness.
Listed as beginner, though the first chapter is a deep learning review, so it helps to bring some background. The description credits Dan Hendrycks as the developer.
ML systems are rapidly increasing in size, are acquiring new capabilities, and are increasingly deployed in high-stakes settings. As with other powerful technologies, safety for ML should be a leading research priority. In this course we’ll discuss how researchers can shape the process that will lead to strong AI systems and steer that process in a safer direction. We’ll cover various technical topics to reduce existential risks (X-Risks) from strong AI, namely withstanding hazards (“Robustness”), identifying hazards (“Monitoring”), reducing inherent ML system hazards (“Alignment”), and reducing systemic hazards (“Systemic Safety”). At the end, we will zoom out and discuss additional abstract existential hazards and discuss how to increase safety without unintended side effects.
🔗 See https://course.mlsafety.org for more.
🔗 Website for the Center for AI Safety: https://www.safe.ai/
🔗 Their newsletter: https://newsletter.safe.ai/
✏️ Dr. Dan Hendrycks developed this course.
❤️ Try interactive AI courses we love, right in your browser: https://scrimba.com/freeCodeCamp-AI (Made possible by a grant from our friends at Scrimba)
⭐️ Contents ⭐️
(0:00:00) Introduction
(0:11:09) Deep Learning Review
(0:52:41) Risk Decomposition
(1:06:57) Accident Models
(1:39:22) Black Swans
(1:58:45) Adversarial Robustness
(2:29:40) Black Swan Robustness
(2:52:56) Anomaly Detection
(3:35:32) Interpretable Uncertainty
(3:59:09) Transparency
(4:12:22) Trojans
(4:22:52) Detecting Emergent Behavior
(4:43:07) Honest Models
(5:00:06) Machine Ethics
(5:52:08) ML for Improved Decision-Making
(6:04:40) ML for Cyberdefense
(6:25:00) Cooperative AI
(6:58:33) X-Risk Overview
(7:05:23) Possible Existential Hazards
(7:13:16) AI and Evolution
(8:03:08) Safety-Capabilities Balance
(8:21:07) Review and Conclusion
🎉 Thanks to our Champion and Sponsor supporters:
👾 davthecoder
👾 jedi-or-sith
👾 南宮千影
👾 Agustín Kussrow
👾 Nattira Maneerat
👾 Heather Wcislo
👾 Serhiy Kalinets
👾 Justin Hual
👾 Otis Morgan
--
Learn to code for free and get a developer job: https://www.freecodecamp.org
Read hundreds of articles on programming: https://freecodecamp.org/news
Advertisement
No Korshub reviews yet.
Ethics and Governance in the Age of Generative AI
