Algorithms we Live by

Instructor: Pantelis Pipergias Analytis

Meeting time and location:
Monday 2 - 4:25 PM
Upson Hall 102

Office hours:
Monday 10 - 11 AM
Gates Hall 223

Course number:
Special topics in Information Science - INFO 4940

Announcement board and forum:
https://piazza.com/configure-classes/spring2018/info4940

Teaser: Artificial intelligence (AI) has become an integral part of our lives. Artificial agents make more and more decisions for us, but we often treat them as black boxes and we are unaware of their inner workings. Still, in most cases artificial agents solve the very same problems that humans and other living organisms have evolved to cope with for their subsistence. What’s more, living intelligence has provided ample inspiration when AI research was leaving the cradle. Neural networks, inspired from the functioning of the human brain, and reinforcement learning, initially developed to describe animal learning, were embraced and further developed by AI researchers to the extent that they currently define the state-of-the-art for tackling many challenging AI problems. Nowadays, frameworks developed to support artificial agents making decisions can be employed to shed new light on how humans make decisions.

Description: In this course we will examine human and artificial intelligence comparatively. We will look at state-of-the-art cognitive models that describe how humans cope with a wide range of demanding tasks. Then we will use that knowledge as a lever to understand machine learning models for dealing with the same problems (or vice versa). The topics discussed will include, among others, categorization and forecasting, recommendation, ranking and search, optimization and planning, creativity, as well as different flavors of multi-armed bandits and reinforcement learning. We will examine scenarios where humans are assisted by artificial intelligence, and problems where humans compete against algorithms. Further, we will discuss whether human cognitive processes can be recovered from the behavioral traces that people leave behind on the Internet; Finally, we will look at problems where human intelligence can provide insights and inspiration for the development of new methods in the quest for better AI agents.

Course objectives: By the end of the course students are expected to develop a better understanding of human cognition and machine learning algorithms. They should be able to build cognitive models to describe human behavior and machine learning models for solving practical everyday life problems such as categorization, estimation and ranking.

Pre-requisites

Math - Familiarity with basic calculus, introductory probability and linear algebra will help to quickly absorb the material and will be useful in the homework assignments. Programming - In some of the assignments you will be asked to program cognitive and machine learning models. Some familiarity with programming will help you get to speed quickly, but is not a requirement for the course.

Grading

The final grade will be based on a mid-term and a final exam (25 % each), a semester project (25 %), three homework assignments (5 % each) and class participation (10%).

Grading rubric: A = 92-100, B=82-88, C=72-78, D=62-68, F = below 60.

Recommended general readings

Overview of topics and schedule

Detailed resources and materials for the class