当前位置:网站首页>MIT-6874-Deep Learning in the Life Sciences Week 7
MIT-6874-Deep Learning in the Life Sciences Week 7
2022-07-05 06:00:00 【Wooden girl】
Lecture 05 Interpretable Deep Learning
- Interpretable deep learning
- One 、Intro to Interpretability
- 1a. Interpretability definition: Convert implicit NN information to human-interpretable information
- 1b. Motivation: Verify model works as intended; debug classifier; make discoveries; Right to explanation
- 1c. Ante-hoc (train interpretable model) vs. Post-hoc (interpret complex model; degree of “locality”)
- 2. Interpreting Deep Neural Networks
- Evaluating Attribution Methods
- 3a. Qualitative: Coherence: Attributions should highlight discriminative features / objects of interest
- 3b. Qualitative: Class Sensitivity: Attributions should be sensitive to class labels
- 3c. Quantitative: Sensitivity: Removing feature with high attribution --> large decrease in class probability
- 3d. Quantitative: ROAR & KAR. Low class prob cuz image unseen --> remove pixels, retrain, measure acc. drop
Interpretable deep learning
This section discusses the explicability of deep learning . Model itself means knowledge , Explicability is of great importance to people like deep learning “ Black box model ” for , It is the root of explaining the reason and method why he made such a judgment , It can help the model work in the direction of human expectations . In many scenes , If recommended 、 Medical and other scenarios have great application prospects .
The following is the outline of this class
One 、Intro to Interpretability
1a. Interpretability definition: Convert implicit NN information to human-interpretable information
1b. Motivation: Verify model works as intended; debug classifier; make discoveries; Right to explanation
Why Interpretability?
- 1.Verify that model works as expected: Wrong decisions can be costly and dangerous
- 2. Improve / Debug classifier
- 3. Make new discoveries
- 4.Right to explanation
“Right to be given an explanation for an output of the algorithm”
1c. Ante-hoc (train interpretable model) vs. Post-hoc (interpret complex model; degree of “locality”)
Know good writing : Explainable AI The research of
Ex post explanation VS Self explanation
There are two ways to obtain the interpretability of the model ( Interpretive classification )
- Ante-hoc & Post-hoc
1. Ante-hoc Interpretability —— Prior interpretability ( Model built-in interpretability )
By training a self explanatory model , To get an explanation of the results .
Common interpretable models :
- Naive Bayes
- Linear regression
- Decision tree
- Rule based models
But the complexity of this kind of model is limited , As a result, its fundamental performance is limited
2. Post-hoc Interpretability—— Ex post interpretability
Refers to the black box model , By some means , Reflect his decision logic
Several levels of interpretability :
- Model level explicability :DNN Why does the model decide the decision boundary so
- Interpretability of features : Which features can maximize the activation of the current model
- Towards individual explicability : Explain why this input is so classified
2. Interpreting Deep Neural Networks
2a. Interpreting Models (macroscopic, understand internals) vs. decisions (microscopic, practical applications)
( Course trend ) Several classifications of model interpretability
- Interpreting decisions:
- Attribution method: What attributes determine the current output of the model
- Example-based: What special case leads to the current output of the model
- Interpreting models:
- Representation analysis: The model represents itself
- Data generation: How to use models to generate data
- Example-based: Related cases
DNN interpretability It can be divided into macro and micro levels
Interpreting models It can be divided into the following four aspects , among The analysis of representation can be divided into weight visualization and proxy model
2b. Interpreting Models: Weight visualization, Surrogate model, Activation maximization, Example-based
1. Weight visualization Weight Visualization
Yes CNN Each layer of filter is visualized , To understand what the model is learning at the current level
2. Surrogate model Agent model
Use a simple ,“ Explicable ” Model to “summarize” Model Output , Try to explain “black box” Output .
3. Data Generation / Activation maximization The data generated / Activate maximize
Activate maximize : Find the way to maximize the activation of neurons , Find the input X, Maximize the probability of the model under the current category
Convolution and deconvolution of the model
The initial input is chaotic , With the increase of training layers , Gradually, we can distinguish the characteristics between numbers
advantage : The advantages and disadvantages of this approach
- DNN It can be explained by looking for the input mode that maximizes the output .
- Connecting with data can improve the interpretability of visualization .
4. Example-based
Summary :
- By visualizing the weight of each layer
- Replace models with low accuracy but strong interpretability
- To some extent, we can obtain useful features by maximizing the activation function / Information
- Through effective construction prototype and criticism, Guide model learning to obtain the most useful , Information for distinguishing
2c. Interpreting Decisions:
Example-based
The input training samples have a decisive impact on the results of the model
Attribution Methods: why are gradients noisy?
Give each pixel a causal score , That is, how much the current pixel contributes to the model to this result .
The result of visualizing attribution
The key is , structure ( features ) Saliency map (Saliency Map)
promote saliency map Methods , First : Change of thinking
hypothesis 1: saliency map Is real
- Some pixels randomly distributed in the image are crucial to how the network makes decisions .
- Noise is very important
hypothesis 2: The gradient is discontinuous - DNN Use piecewise linear functions (ReLU Activate ,max-pooling etc. ).
- The mutation jump of importance score on the infinitesimal change of input .
hypothesis 3: - A feature may have a strong impact on a global scale , But it will have a small impact locally
Other attribution methods
Gradient-based Attribution: SmoothGrad, Interior Gradient
Backprop-based Attribution: Deconvolution, Guided Backpropagation
- Observe : Removing more gradients will bring a clearer visual effect
Evaluating Attribution Methods
3a. Qualitative: Coherence: Attributions should highlight discriminative features / objects of interest
- Attribution should be based on distinctive characteristics
3b. Qualitative: Class Sensitivity: Attributions should be sensitive to class labels
- Attribution should be category sensitive
3c. Quantitative: Sensitivity: Removing feature with high attribution --> large decrease in class probability
- Removing features with high attributes will lead to a significant reduction in class probability
3d. Quantitative: ROAR & KAR. Low class prob cuz image unseen --> remove pixels, retrain, measure acc. drop
边栏推荐
- Mysql database (I)
- Daily question 2013 Detect square
- Fried chicken nuggets and fifa22
- Implement a fixed capacity stack
- Analysis of backdoor vulnerability in remote code execution penetration test / / phpstudy of national game title of national secondary vocational network security B module
- 【Rust 笔记】15-字符串与文本(上)
- PC寄存器
- Dynamic planning solution ideas and summary (30000 words)
- 【Jailhouse 文章】Performance measurements for hypervisors on embedded ARM processors
- [practical skills] technical management of managers with non-technical background
猜你喜欢
[practical skills] technical management of managers with non-technical background
Implement an iterative stack
Introduction et expérience de wazuh open source host Security Solution
剑指 Offer 06.从头到尾打印链表
Pointnet++ learning
Educational Codeforces Round 116 (Rated for Div. 2) E. Arena
wordpress切换页面,域名变回了IP地址
CF1634 F. Fibonacci Additions
【Jailhouse 文章】Performance measurements for hypervisors on embedded ARM processors
Brief introduction to tcp/ip protocol stack
随机推荐
Introduction and experience of wazuh open source host security solution
【云原生】微服务之Feign自定义配置的记录
PC register
Full Permutation Code (recursive writing)
剑指 Offer 53 - I. 在排序数组中查找数字 I
Wazuh开源主机安全解决方案的简介与使用体验
网络工程师考核的一些常见的问题:WLAN、BGP、交换机
Pointnet++ learning
【Jailhouse 文章】Look Mum, no VM Exits
Solution to the palindrome string (Luogu p5041 haoi2009)
After setting up the database and website When you open the app for testing, it shows that the server is being maintained
shared_ Repeated release heap object of PTR hidden danger
LeetCode 1200.最小绝对差
CPU内核和逻辑处理器的区别
Educational codeforces round 109 (rated for Div. 2) C. robot collisions D. armchairs
In this indifferent world, light crying
Sword finger offer 05 Replace spaces
1.14 - 流水线
Codeforces Round #716 (Div. 2) D. Cut and Stick
LeetCode 0107.二叉树的层序遍历II - 另一种方法