Главная
Study mode:
on
1
Intro
2
walter Pitts
3
Imagenet
4
The alignment problem
5
Machine learning systems
6
Training data
7
Labels in the wild
8
Representation and data sets
9
What can we do about this
10
TCAV
11
Fire Truck
12
Robustness
13
Open Category Problem
14
Examples
15
What makes a system fair
16
Predicting reoffending
17
Ground truth
18
Potential ethical issue
19
Criminal justice
20
Risk vs Needs
21
Reinforcement Learning
22
Facebook Reinforcement Learning
23
Coast Runners 3
24
Sparsity
25
Designing reward functions
26
Behavior cloning
27
Shadow mode
28
Inverse reinforcement learning
Description:
Explore the complexities and ethical challenges of machine learning in this thought-provoking lecture by Brian Christian. Delve into the alignment problem, examining how AI systems can fail to meet human expectations and the potential risks that arise. Learn about issues in training data, representation in datasets, and the open category problem. Investigate fairness in machine learning, particularly in criminal justice applications, and understand the intricacies of reinforcement learning. Discover innovative approaches like TCAV, behavior cloning, and inverse reinforcement learning that aim to address these challenges. Gain insights into the ongoing efforts to create AI systems that better align with human values and expectations.

How Can Machines Learn Human Values? - With Brian Christian

The Royal Institution
Add to list
0:00 / 0:00