Difference between revisions of "DSC Capstone2020"

From Immersive Visualization Lab Wiki
Jump to: navigation, search
(→‎Papers)
 
(60 intermediate revisions by one user not shown)
Line 1: Line 1:
−
=DSC 180 Capstone Domain: Explainable AI (Section A01)=
+
=DSC 180A Capstone Section A01: Explainable AI=
  
 
* Instructor: [https://jacobsschool.ucsd.edu/faculty/profile?id=360 Jurgen Schulze]
 
* Instructor: [https://jacobsschool.ucsd.edu/faculty/profile?id=360 Jurgen Schulze]
 
* Discussion: Wednesdays 12-12:50pm on Zoom at https://ucsd.zoom.us/j/9100475160
 
* Discussion: Wednesdays 12-12:50pm on Zoom at https://ucsd.zoom.us/j/9100475160
 +
* Office hours: Fridays 10-11am on Zoom at https://ucsd.zoom.us/j/97761107672
 
* Piazza Discussion Board: https://piazza.com/ucsd/fall2020/dsc180
 
* Piazza Discussion Board: https://piazza.com/ucsd/fall2020/dsc180
  
Line 9: Line 10:
 
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.
 
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 go over the basics of CNNs, then learn about saliency and attention maps, and finally we are going to implement the [http://gradcam.cloudcv.org Grad-CAM algorithm] in [https://pytorch.org PyTorch] and apply it to the [https://cocodataset.org COCO image data set].
+
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 [http://gradcam.cloudcv.org Grad-CAM algorithm] in [https://pytorch.org PyTorch] and apply it to the [https://cocodataset.org 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.
 
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.
 
This class will be entirely remote.
 +
 +
You can use the [https://dsc-capstone.github.io/resources/computing/ DSMLP cluster], which already has the COCO data set installed.
  
 
==Schedule==
 
==Schedule==
Line 20: Line 23:
 
|-
 
|-
 
! Week
 
! Week
−
! Date
+
! Discussion Date
−
! Discussion Topic
+
! Discussion Wed 12-12:50pm (Link to Slides)
−
! Replication Tasks
+
! Participation Tasks (Due Tuesdays at noon)
 +
! Checkpoints
 
|-  
 
|-  
 
| 1
 
| 1
 
| Oct 7
 
| Oct 7
−
| Overview
+
| [[Media:01_CourseOverviewF20.pdf|Overview]]
 
| None for week 1
 
| None for week 1
 +
|
 
|-
 
|-
 
| 2
 
| 2
 
| Oct 14
 
| Oct 14
−
|  
+
| [[Media:02_SaliencyMapsF20.pdf|Saliency Maps]]
−
| Literature review
+
| [[DSC180F20W2|Literature review]] (due Oct 13)
 +
|
 
|-
 
|-
 
| 3
 
| 3
 
| Oct 21
 
| Oct 21
−
|  
+
| [[Media:03_CocoDatasetF20.pdf|COCO Image Data Set]]
−
| Description of data
+
| [[DSC180F20W3|COCO Dataset]] (due Oct 20)
 +
|
 
|-
 
|-
 
| 4
 
| 4
 
| Oct 28
 
| Oct 28
−
|  
+
| [[Media:04_Grad-CAMF20.pdf|Introduction to Grad-CAM]]
−
| Checkpoint #1 due
+
| [[DSC180F20W4|Custom COCO Demo]] (due Oct 27)
 +
| [[DSC180F20W4|Replication Checkpoint #1]] (due Oct 30)
 
|-
 
|-
 
| 5
 
| 5
 
| Nov 4
 
| Nov 4
−
|  
+
| [[Media:05_DeepLearningF20.pdf|Deep Learning]]
−
| Description of methods
+
| [[DSC180F20W5|Introduction to Grad-CAM]] (due Nov 3)
 +
|
 
|-
 
|-
 
| 6
 
| 6
 
| Nov 11
 
| Nov 11
 
| Veterans Day (No Discussion)
 
| Veterans Day (No Discussion)
−
| Implementation of methods
+
| [[DSC180F20W6|Deep Learning]] (due Nov 10)
 +
|
 
|-
 
|-
 
| 7
 
| 7
 
| Nov 18
 
| Nov 18
−
|  
+
| [[Media:06_GradCAM_in_ContextF20.pdf|Grad-CAM in Context]]
−
| Checkpoint #2 due
+
| [[DSC180F20W7P|PyTorch]] (due Nov 17)
 +
| [[DSC180F20W7|Replication Checkpoint #2]] (due Nov 22)
 
|-
 
|-
 
| 8
 
| 8
 
| Nov 25
 
| Nov 25
−
|  
+
| [[Media:07_GradCAM_ForXAI_F20.pdf|Grad-CAM for XAI]]
−
| Implementation of Grad-CAM
+
| [[DSC180F20W8P|Grad-CAM in Context]] (due Nov 24)
 +
|
 
|-
 
|-
 
| 9
 
| 9
 
| Dec 2
 
| Dec 2
−
|  
+
| [[Media:08_Grad-CAM_F20.pdf|Grad-CAM and Q2 Project]]
−
| Implementation of Grad-CAM
+
| [[DSC180F20W9P|Grad-CAM for XAI]] (due Dec 1)
 +
|
 
|-
 
|-
 
| 10
 
| 10
 
| Dec 9
 
| Dec 9
−
|  
+
| [[Media:09_ElevatorPitches_F20.pdf|Q2 Project Elevator Pitches]]
−
| Final Report due
+
| [[DSC180F20W10P|Grad-CAM Evolution]] (due Dec 8)
 +
| [[DSC180F20W10|Final Replication Report]] (due Dec 11)
 
|}
 
|}
  
−
==Papers==
+
==Relevant Publications==
  
 
* Paper for replication: [https://arxiv.org/pdf/1610.02391v1.pdf Grad-CAM: Why did you say that? Visual Explanations from Deep Networks via Gradient-based Localization]
 
* Paper for replication: [https://arxiv.org/pdf/1610.02391v1.pdf Grad-CAM: Why did you say that? Visual Explanations from Deep Networks via Gradient-based Localization]
 
* Paper on COCO image dataset: [https://arxiv.org/abs/1405.0312 Microsoft COCO: Common Objects in Context]
 
* Paper on COCO image dataset: [https://arxiv.org/abs/1405.0312 Microsoft COCO: Common Objects in Context]
 +
* Original attention map paper: [https://arxiv.org/pdf/1312.6034.pdf Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps]
 
* [https://arxiv.org/pdf/1512.04150.pdf Learning Deep Features for Discriminative Localization]
 
* [https://arxiv.org/pdf/1512.04150.pdf Learning Deep Features for Discriminative Localization]
 
* [https://arxiv.org/pdf/1311.2901.pdf Visualizing CNNs with deconvolution]
 
* [https://arxiv.org/pdf/1311.2901.pdf Visualizing CNNs with deconvolution]
 +
* [http://faculty.neu.edu.cn/yury/AAI/Textbook/Deep%20Learning%20with%20Python.pdf Deep Learning with Python by François Chollet]
  
 
==Useful Links==
 
==Useful Links==

Latest revision as of 09:50, 6 January 2021

Contents

DSC 180A 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 Discussion Date Discussion Wed 12-12:50pm (Link to Slides) Participation Tasks (Due Tuesdays at noon) Checkpoints
1 Oct 7 Overview None for week 1
2 Oct 14 Saliency Maps Literature review (due Oct 13)
3 Oct 21 COCO Image Data Set COCO Dataset (due Oct 20)
4 Oct 28 Introduction to Grad-CAM Custom COCO Demo (due Oct 27) Replication Checkpoint #1 (due Oct 30)
5 Nov 4 Deep Learning Introduction to Grad-CAM (due Nov 3)
6 Nov 11 Veterans Day (No Discussion) Deep Learning (due Nov 10)
7 Nov 18 Grad-CAM in Context PyTorch (due Nov 17) Replication Checkpoint #2 (due Nov 22)
8 Nov 25 Grad-CAM for XAI Grad-CAM in Context (due Nov 24)
9 Dec 2 Grad-CAM and Q2 Project Grad-CAM for XAI (due Dec 1)
10 Dec 9 Q2 Project Elevator Pitches Grad-CAM Evolution (due Dec 8) Final Replication Report (due Dec 11)

Relevant Publications

Useful Links

Direct CNN Visualization

Retrieved from "http://ivl.calit2.net/wiki/index.php?title=DSC_Capstone2020&oldid=13646"