When: Tu/Th 4:30 pm -- 5:45 pm.
Where: LWSN 1106.
Instructor: Yexiang Xue, yexiang@purdue.edu.
Office Hour: Mon. 4:30 pm -- 5:45 pm. Where: LWSN 2142V or 1187.
Online discussion is available at Blackboard (mycourses.purdue.edu).
Balancing environmental, economic, and societal needs for a sustainable future encompasses problems of unprecedented size and complexity. Artificial intelligence can play an important role in addressing critical sustainability challenges faced by present and future generations. The goal of this course has two folds. First, this course intends to expose students with fundamental computational tools to address sustainability challenges, such as mathematical programming, planning, constraint satisfaction, multi-agent modeling and statistical machine learning. Second, this course intends to motivate students with successful applications of artificial intelligence on real-world problems in sustainability. We intend to cover successful applications of artificial intelligence in discovering new materials with human computation, ecological monitoring through citizen science programs, poverty mapping with deep learning, stopping illegal poaching with game theory, etc.
Classes will consist of instructor presentations, student presentations, and group discussions. The first few lectures consist of introductions to basic computational tools, such as constraint programming, probabilistic inference, supervised and unsupervised learning, etc. Then the course moves to discussing successful applications of AI on sustainability-related fields. Students are expected to (1) read, discuss, and present research papers, (2) complete a semester-long class project in groups, (3) review and comment on one class project proposal from another group.
Basic knowledge of linear algebra and calculus, a basic course in probability and statistics (e.g., STAT301/350/416, CS373), and basic programming skills (e.g., CS381) are required. Students without this background should have a discussion with the instructor.
Senior undergraduate or graduate students with interest in machine learning, data mining and Artificial Intelligence in general. This course also welcomes students from related fields, such as agriculture, economics, applied math, physics, chemistry, and engineering, who are interested in using computational tools to solve real-world problems in their domain of interest.
| Time | Topic | Notes |
|---|---|---|
| 8/21 Tues. | Introduction
Slides and annoucements will be posted on Blackboard (mycourses.purdue.edu)! Please let the instructor know if you cannot log into the course page on Blackboard. |
|
| 8/23 Thurs. | AI in Statistical Physicists' Perspective: phase transition; heavy-tail phenomenon; random restarts. Can you imagine that AI problems have a "melting" temperature, just like ice? Why is it not always a good idea to follow the gradients? Reading: Critical Behavior in the Satisfiability of Random Boolean Expressions. [pdf]. Can Get Satisfaction.[pdf]. Optional Reading: Pinning Down a Treacherous Border in Logical Statements.[pdf]. Critical Behavior in the Computational Cost of Satisfiability Testing.[pdf]. Generating Hard Satisfiability Problems [pdf]. Heavy-tailed Distributions in Combinatorial Search. [pdf]. Heavy-Tailed Phenomena in Satisfiability and Constraint Satisfaction Problems. [pdf]. |
|
| 8/28 Tues. | Invited talk: Peek into On-demand Economy. | |
| 8/30 Thurs. | AI in Statistical Physicists' Perspective | |
| 9/4 Tues. | Statistical Relational Learning: Bayesian networks; Markov logic network; conditional random field. History progressed from logic to probability. |
|
| 9/6 Thurs. | Statistical Relational Learning | |
| 9/11 Tues. | Probabilistic Inference: variational approach; sampling. How to make inference in a statistical relational model? |
|
| 9/13 Thurs. | Probabilistic Inference: XOR stream-lining. Why is it not always a good idea to do MCMC? |
|
| 9/18 Tues. | Unsupervised Learning: contrasive divergence. Why is probabilitic inference a central piece in unsupervised learning? Reading: Ian Goodfellow, Yoshua Bengio, Aaron Courville, Deep Learning. chapter 18: Confronting the Partition Function. [pdf]. Miguel A. Carreira-Perpinan, Geoffrey Hinton, On Contrastive Divergence Learning. [pdf]. |
|
| 9/20 Thurs. | Unsupervised Learning: deep probabilistic embedding; generative adversarial networks. How is what we have talked about connected to recent progress in deep unsupervised learning? Reading: Ian J. Goodfellow et. al. Generative Adversarial Nets. [pdf]. |
|
| 9/25 Tues. | Poverty Mapping with Satellite Imagery How can remote-sensing help reduce poverty rate? Reading: Neal Jean, et. al. Combining Satellite Imagery and Machine Learning to Predict Poverty. [pdf]. |
|
| 9/27 Thurs. | Poverty Mapping with Satellite Imagery
Reading: Luming Tang, Yexiang Xue, Di Chen, Carla P. Gomes, Multi-Entity Dependence Learning with Rich Context via Conditional Variational Auto-encoder. [pdf]. |
|
| 10/2 Tues. | Wildlife Corridor Design How is AI useful in maintaining landscape connectivity for wild animals? Reading: Bistra Dilkina, Katherine Lai, Carla P. Gomes. Upgrading Shortest Paths in Networks. [pdf]. |
|
| 10/4 Thurs. | Wildlife Corridor Design
Reading: Daniel Sheldon, et. al. Maximizing the Spread of Cascades Using Network Design. [pdf]. |
|
| 10/9 Tues. | No class | |
| 10/11 Thurs. | Games in Citizen Science How can game theory help reduce the data bias in citizen science? Reading: Yexiang Xue, Ian Davies, Daniel Fink, Carla P. Gomes. Avicaching: A Two Stage Game for Bias Reduction in Citizen Science. [pdf]. |
|
| 10/16 Tues. | Games in Citizen Science
Reading: Yexiang Xue, Ian Davies, Daniel Fink, Carla P. Gomes. Behavior Identification in Two-stage Games for Incentivizing Citizen Science Exploration. [pdf]. |
|
| 10/18 Thurs. | Green Security Games Game theory combats illegal poaching. Reading: Fei Fang, Peter Stone, Milind Tambe. When Security Games Go Green: Designing Defender Strategies to Prevent Poaching and Illegal Fishing. [pdf]. |
|
| 10/23 Tues. | Green Security Games
Reading: Chun Kai Ling, Fei Fang, J. Zico Kolter. What game are we playing? End-to-end learning in normal and extensive form games. [pdf]. |
|
| 10/25 Thurs. | Pareto Frontier in Hydropower Dam Placement Trade-off of multiple societal and economical objectives in the Amazon river basin. Reading: Christos Papadimitriou, Mihalis Yannakakis. On the Approximability of Trade-offs and Optimal Access of Web Sources. [pdf]. |
|
| 10/30 Tues. | Pareto Frontier in Hydropower Dam Placement
Reading: Xiaojian Wu, et. al. Efficiently Approximating the Pareto Frontier: Hydropower Dam Placement in the Amazon Basin. [pdf]. |
|
| 11/1 Thurs. | AI for Materials Discovery Crowdsouring, human computation acclerates the pace of materials discovery. Reading: Yexiang Xue, et. al. Phase-Mapper: An AI Platform to Accelerate High Throughput Materials Discovery. [pdf]. |
|
| 11/6 Tues. | AI for Materials Discovery
Reading: Ronan Le Bras, et. al. Crowdsourcing Backdoor Identification for Combinatorial Optimization. [pdf]. |
|
| 11/8 Thurs. | Species Distribution Modeling Does deep learning help to model where animals live? Reading: Daniel Fink, Theodoros Damoulas, Jaimin Dave, Adaptive Spatio-Temporal Exploratory Models: Hemisphere-Wide Species Distributions from Massively Crowdsourced eBird Data. [pdf]. Daniel Fink, et. al. Spatiotemporal exploratory models for broad-scale survey data. [pdf]. |
|
| 11/13 Tues. | Species Distribution Modeling
Reading: Kevin Winner, Daniel Sheldon. Probabilistic Inference with Generating Functions for Poisson Latent Variable Models. [pdf]. |
|
| 11/15 Thurs. | Stochastic Optimization Optimal decision making guaranteed on probabilistic models learned from data. Reading: Anton J. Kleywegt, Alexander Shapiro, and Tito Homem-de Mello. The sample average approximation method for stochastic discrete optimization. [pdf]. |
|
| 11/20 Tues. | Stochastic Optimization
Reading: Yexiang Xue, Zhiyuan Li, Stefano Ermon, Carla P. Gomes, Bart Selman. Solving Marginal MAP Problems with NP Oracles and Parity Constraints. [pdf]. |
|
| 11/22 Thurs. | Thanksgiving. No class. | |
| 11/27 Tues. | Critical Behavior in Random Graphs and Satisfiability. Mathematical explanations of the critical behavior. Reading: Chapter 8.1 -- 8.5 in Avrim Blum, John Hopcroft, and Ravindran Kannan. Foundations of Data Science. [pdf]. |
|
| 11/29 Thurs. | Critical Behavior in Random Graphs and Satisfiability.
Reading: Chapter 8.6 -- 8.10 in Avrim Blum, John Hopcroft, and Ravindran Kannan. Foundations of Data Science. [pdf]. |
|
| 12/4 Tues. | Final project presentation. | |
| 12/6 Thurs. | Final project presentation. |
Synthetic and real datasets for materials discovery.
Dataset for the corridor-design problem and landscape optimization problem.
Remote sensing images (a code repository which contains code to download from Google Earth engine).