Software Development

AI Safety – Full Course by Safe.AI Founder on Machine Learning & Ethics (Center for AI Safety)



ML techniques are quickly growing in dimension, are buying new capabilities, and are more and more deployed in high-stakes settings. As with different highly effective applied sciences, security for ML needs to be a number one analysis precedence. On this course we’ll focus on how researchers can form the method that may result in sturdy AI techniques and steer that course of in a safer path. We’ll cowl numerous technical matters to scale back existential dangers (X-Dangers) from sturdy AI, particularly withstanding hazards (“Robustness”), figuring out hazards (“Monitoring”), decreasing inherent ML system hazards (“Alignment”), and decreasing systemic hazards (“Systemic Safety”). On the finish, we are going to zoom out and focus on extra summary existential hazards and focus on the best way to improve security with out unintended uncomfortable side effects.

Thank you for reading this post, don't forget to subscribe!

🔗 See for extra.
🔗 Web site for the Center for AI Safety:
🔗 Their e-newsletter:

✏️ Dr. Dan Hendrycks developed this course.

⭐️ Contents ⭐️
(0:00:00) Introduction
(0:11:09) Deep Learning Evaluation
(0:52:41) Danger Decomposition
(1:06:57) Accident Fashions
(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 Habits
(4:43:07) Sincere Fashions
(5:00:06) Machine Ethics
(5:52:08) ML for Improved Choice-Making
(6:04:40) ML for Cyberdefense
(6:25:00) Cooperative AI
(6:58:33) X-Danger Overview
(7:05:23) Doable Existential Hazards
(7:13:16) AI and Evolution
(8:03:08) Safety-Capabilities Steadiness
(8:21:07) Evaluation and Conclusion

🎉 Due 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

Study to code without spending a dime and get a developer job:

Learn lots of of articles on programming:

source