Difference between revisions of "DSC Capstone2020"

From Immersive Visualization Lab Wiki
Jump to: navigation, search
(Schedule)
(Schedule)
Line 67: Line 67:
 
| Nov 18
 
| Nov 18
 
|  
 
|  
| [[DSC180F20W7|Replication Checkpoint #2]] due
+
|
| Replication Checkpoint #2
+
| [[DSC180F20W7|Replication Checkpoint #2]]
 
|-
 
|-
 
| 8
 
| 8
Line 85: Line 85:
 
| Dec 9
 
| Dec 9
 
|  
 
|  
| [[DSC180F20W10|Final Replication Report]] due, Capstone Project proposal due
+
| Capstone Project proposal
| Final Replication Report
+
| [[DSC180F20W10|Final Replication Report]]
 
|}
 
|}
  

Revision as of 11:29, 28 October 2020

Contents

DSC 180 Capstone Section A01: Explainable AI

Overview

In this capstone domain we are going to study how we can make machine learning systems more user friendly by exploiting additional knowledge we can derive from the system and present it to the user. These types of systems are called Explainable AI.

The example we are going to use in this class is object recognition in images. We are first going to learn about saliency and attention maps, then get to know a large publicly available image data set (COCO) and finally we are going to implement the Grad-CAM algorithm in PyTorch and apply it to the COCO image data set.

The quarter will end with a proposal for your capstone project, which you will be working on in the winter quarter.

This class will be entirely remote.

You can use the DSMLP cluster, which already has the COCO data set installed.

Schedule

Week Date Discussion Topic Participation Tasks (Due Tuesdays at noon) Checkpoints (Due Fridays)
1 Oct 7 Overview None for week 1
2 Oct 14 Saliency Maps Literature review
3 Oct 21 COCO Image Data Set COCO Dataset
4 Oct 28 Custom COCO Demo Replication Checkpoint #1
5 Nov 4 Description of methods
6 Nov 11 Veterans Day (No Discussion) Implementation of methods
7 Nov 18 Replication Checkpoint #2
8 Nov 25 Implementation of Grad-CAM
9 Dec 2 Implementation of Grad-CAM
10 Dec 9 Capstone Project proposal Final Replication Report

Papers

Useful Links

Direct CNN Visualization