CS/INFO 6702 Topics in Computational Sustainability

 

Computer Science and Information Science

Spring 2010

Cornell University

 

Instructor: Carla Gomes

 

Faculty Team: Jon Conrad,  Steve Ellner, Carla Gomes, and  Mary Lou Zeeman

 

Teaching Assistants: Bistra Dilkina and Georgios Piliouras

 

Time: WF 1:25-2:40 pm.

 

Location:   Snee Hall

 

Office Hours: Wednesday 3:15-4:30 PM (5133 Upson Hall).

 

Web page: http://www.cs.cornell.edu/Courses/cs6702/2010sp

 

 

Course Work

 

The course work consists of three components:

1.    Attendance and participation in the lectures

2.    A reaction paper on a particular computational sustainability topic, a presentation of a research problem in class, or a good annotated bibliography.

3.    A final project, including an initial project proposal.

Grade option: 1, 2, and 3 required.

S/U option: 1 and 2 required.

Students are encouraged to work in interdisciplinary groups.  

 

Reaction Paper

The reaction papers are meant to identify and discuss one or two interesting computational research questions concerning a certain sustainability topic. The reaction paper should be around 5 pages in length.

The reaction paper is due on March 3rd, and it is supposed to be individual work.

Instructions for the reaction paper can be found here (or can be downloaded as pdf).

Project

The selection of the topic and scope of the final project is mainly up to the student(s). Projects can be done in groups of size 1-3, or even 4 if you have enough work for everybody. 

A short project proposal (2 pages) briefly outlining the project is required. The project proposal should provide background work and a high level plan for the project. It’s okay to leverage from the reaction paper if the project is an extension of the reaction paper. In that case the proposal should outline how to extend the ideas in the reaction paper or how to address in detail an idea presented in the reaction paper. The project proposal is due on March 17th

There will be a presentation of the projects in class by each group during the last two weeks of classes. The exact schedule for the project presentations will be worked out later in the semester.

A final project report is required. The final report should be an 8-12 page paper, describing the problem, the approach, the results, and related work. The final project write up is due on May 10.

A final poster is given as an option to replace the final report. First draft with preliminary results is due on May 3rd, final version with complete results and Notes that explain the poster is due on May 10.

Here are some examples of posters:

Nash Equilibria in Graphical Games  (AAAI 2007)  PDF

Optimization Models for Red-Cockaded Woodpecker Management (CompSust 2009) PDF

Learning with Resource Capacity Constraints: Pastoralists’ Mobility In Kenya (CS Visit Day)   PDF   

Computational Thinking for Material Discovery: Bridging Constraint Reasoning and Learning (CS Visit Day)   PDF

 

 

Here are some different types of projects:

           Programming project (ideally with principled experimentation).

           Experimental evaluation or empirical experiment of a model, algorithm.

           An original dataset describing certain phenomena with a detailed analysis. 

           More research oriented project (perhaps involving programming too).

           Survey paper (has to be original!) on computational issues concerning a sustainability topic.

Here are some examples of projects:

Novel Models:
Biodiversity conservation - study some aspect of biodiversity conservation planning by creating an optimization model/technique with experimental evaluation
Socioeconomic aspects of sustainability - How can economic incentives and sustainability coexist?  How do we address realistic concerns (e.g. discounting of future costs, tragedy of the commons). Mechanism design for conservation or carbon emission credits

Description of a computational sustainability research problem

Data Modeling, simulation, and Analysis:
Statistical/machine learning approaches for time-series spatially explicit data of land cover (for conservation or climate change prediction)
Species Distribution Modeling - Machine learning techniques to obtain more accurate species distribution models from uncertain and missing data (Lab of Ornithology)
Ecosystem Modeling - Population Dynamics in Networks (Co-evolution of Population, Networks)
Modeling of Disease Outbreaks -  (Overlay with Google maps, Identify hotspots)

Analysis of Bibliographic Network:
Social Network Analysis of the Computational Sustainability community - use research paper citations to identify the key papers/people in computational sustainability
Social Network Analysis of the Computational Sustainability research topic - use research paper citations to to track the time series development of the research topic

Computer Games/Applications:
Design a computer game that introduces some computational sustainability concept to kids
Design an iPhone application addressed towards adults but with sustainability overtones (e.g. eco-SimCity)
Design a Facebook game or application that allows individuals to receive social recognition by publicizing their eco-friendliness.
Design a prediction market application for sustainability questions (i.e. predict the highest temperature for the next August)
Design an artificial market for carbon emission credit

Extension of UrbanSim to incorporate a different  computational  model

 

Survey paper:

Critical survey of methodologies to evaluate impacts of biofuels.

Critical survey of quantifying biodiversity.

Critical survey of incentives for CO2 offsetting addressing in particular computational issues.

Critical survey of agent-based models for a particular topic

Critical survey of GIS systems for certain kinds of problems – limitations and opportunities

Critical survey of UrbanSim with

 

 

Academic Integrity

This course follows the Cornell University Code of Academic Integrity. Each student in this course is expected to abide by the Cornell University Code of Academic Integrity. Any work submitted by a student in this course for academic credit will be the student's own work. Violations of the rules (e.g. cheating, copying, non-approved collaborations) will not be tolerated.