纯在折磨这个课 本人期中依托，参考慎重，不知道意义何在
可能会有很多不完善的地方，我也有很多地方感觉一知半解，欢迎讨论和交流。
评论区欢迎补充，需要打开vpn查看
哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈哈

尼玛exam2清朝老题，做往年卷子去吧。。。

Sept. 30th 2023,从今天开始好好记笔记。
开学一个多月，我的评价是快乐不会消失只会转移，一点也乐不出来力。
All of Programming

Ciao!

感谢四年朋友的陪伴

臭狗lzx看这👇

Email : Ping.Zhang@xjtlu.edu.cn
Office : EE222

Email: jie.zhang01@xjtlu.edu.cn
Office hours: 2:00-3:00PM, Tue & Wed EE522

lecture的简单笔记记录

Human-Centric Computing

The Design Life Cycle
It’s a interface/interaction design - an iterative process:
design –> implementation –> evaluation –> design

• not always start with design
  • maybe first figure out what people need/want (evaluation)
  • maybe there was an initial system (implementation)
• quite expensive having to re-implement a (complex) system over and over
• better strategies needed-ideas?
  • different types of “implementation”
  • different degrees of evaluation

为什么要有interface design process?

• to avoid user frustration with the finished product
• to avoid unnecessary costs during product development 减小产品开发过程中不必要的开销
• to avoid additional cost after product development 减小产品开发后的多余开销
• reasons for going over budget
  • user-requested changes
  • overlooked tasks
  • users did not understand their own requirements
  • insufficient user-developer communication and understanding

Foundations for designing interfaces

• Understanding users and their tasks
  • Task-centered system design
    • how to develop task examples
    • how to evaluate designs through a task-centered walk-through
• Designing with the user
  • User-centered design and prototyping
    • methods for designed with the user
    • low and medium fidelity prototyping
• Evaluating interfaces with users
  • the role of evaluation in interface design
  • how to observe people using systems to detect interface problems

Considerations when designing interfaces
do I know who my user and what their tasks are?

• how much computer knowledge?
• novice? occasional user? regular user? expert?
• what do people want to achieve, what are the tasks?

novel interfaces: what hardware I can expect?

Storyline Visualization
Design Principles

• D1 Lines in the same group should appear next to each other.
• D2 Otherwise, lines must be far away from each other.
• D3 A line must remain straight unless its group changes.

Optimization Goals

• reducing line crossings
• reducing line wiggles
• reducing white space

Model representation

Cost function
有时候也会被叫做平方误差代价函数

Gradient descent algorithm 梯度下降法

learning rate后面的部分是derivative
假如阿尔法太小，梯度就会减缓。假如阿尔法太大，梯度下降可能超过最小值。

Machine learning definition
Well-posed learning problem: A computer program is said to learn from experience E with respect to some task T and some performance measure P, if its performance on T, as measured by P, improves with experience E.

Machine learning algorithms:

• supervised learning 我们教计算机做某件事
• unsupervised learning 计算机自己总结学会做某事

Others: Reinforcement learning, recommender systems.

supervised learning
想法是在监督学习中，对于数据集中的每个样本，我们想要算法预测并得到正确答案。
分类问题 classification problem 回归问题 regression problem

unsupervised learning
把大量的数据给到computer 计算机自习学会做某事

1. how do we represent the meaning of a words?
   Definition: meaning (webster dictionary)

• the idea that is represented by a word, phrase, etc.
• and so on.

Commonest linguistic way of thinking of meaning: 最常见的意义思维方式

• signifier <====> signified (idea or thing) = denotation

2. main idea of word2vec

   1	Predict between every word and its context words.

   Two algorithms

• Skip-grams (SG)
  • Predict context words given target (position independent)
• Continuous Bag of Words (CBOW)
  • Predict target word from bag of words context

Two (moderately efficient) training methods

• Hierarchical softmax
• Negative sampling

Details of Word2Vec
Predict surrounding words in a window of radius m of every word.
For p(wt+j|wt) the simple first formulation is

where o is the outside (or output) word index, c is the center word index, vc and uo are “center” and “outside” vectors of indices c and o.
每个词都可以有两个向量（一个也可以 但是两个更简单）

Sentence embedding
Compute sentence similarity using the inner product.

1
2
3
4
5
6
7

S1: Mexico wishes to guarantee citizen's safety.
S2: Mexico wishes to avoid more violence.
Score: 4/5

S1: Iranians Vote in Presidential Election
S2: Keita Wins Mali Presidential Election
Score: 0.4/5

Use as features for sentence classification. 语意感情分析

From Bag-of-words to Complex models

• Bag-of-words BoW

  1	v("natural language processing") = 1/3(v("natural") + v("language") + v("processing"))

• Recurrent neural networks, recursive neural networks, convolutional neural networks…