Lecture Definitions Flashcards
(76 cards)
1
Q
Domain
A
the class of input a function accepts.
2
Q
Range
A
all the possible return values of a function.
3
Q
Abstraction
A
Act of leaving out of consideration one or more properties of a complex object so as to attend to others.
4
Q
Generalisation
A
The process of formulating general concepts by abstracting common properties of instance.
5
Q
Command
A
No outputs, meant for side effects
6
Q
Reporter (function)
A
Any type of output
7
Q
Predicate (function)
A
boolean output
8
Q
function definition
A
- Take in 0 or more inputs ad return only 1 output
- the same inputs MUST yield the same outputs
- Rules: no state, no mutation, no side effects
9
Q
why use functions?
A
- allow for the generalization of code
- they can be composed together to make even more magnificent things
- they allow for functional programming (increased speeds etc)
10
Q
Functional Programming
A
- programming only using functions (not commands)
- as they don’t have state they an run on multiple systems at the same time with allows for faster computing and cloud computing.
11
Q
Recursion
A
Technique for defining functions that use themselves to complete their own definition.
12
Q
Overflow
A
when the number of represented things exceeds the number of digits allocated for it
13
Q
Roundoff
A
True real number can’t be stored exactly (due to fixed number of bits)
14
Q
1 nibble
A
4 bits
15
Q
1 hexadigit
A
16 things
16
Q
8 bits
A
1 byte
17
Q
1 byte
A
8 bits
18
Q
2 hexadgits
A
256 things
19
Q
N bits
A
2^N things
20
Q
Computing tools and techniques…
A
are used to create computational artifacts (e.g. IDEs: integrated development environments)
enhance process of finding a solution to a problem.
21
Q
Effective collaboration teams practice:
A
interpersonal communication, consensus building, conflict resolution, negotiation.
22
Q
Programming Paradigms
A
the concepts and abstractions used to represent the elements of a program and the steps that compose a computation
or:
a way to classify the style of programming
23
Q
functional model of programing
A
Computation as evaluation of functions. (therefore posses rules of functions)
24
Q
Imperative Programming
A
Sequential programming (computation as a series of steps) -rules: allows mutations, allows assignment
25
Is functional style code an example of imperative style code or is imperative style code an example of functional style code?
functional style code is a subset f imperative style code.
26
What is OOP
Object Orientated Programming.
Programming which requires sending of messages between objects to simulate the temporal evolution of a set of real world phenomena.
Objects are data structures which have a local state and methods (behaviors) which you ask of them.
Eg (java, c++)
27
What are Classes and instances?
used in OOP.
Instances are examples of classes e.g. Dan Garcia is instance of Professor.
Inheritance saves code by arranging heirarchical classes e.g. pianist special case of musician etc.
28
Who is Dr Ivan Sutherland?
"Father of computing graphics" who wrote sketchpad
29
Declarative Programming
Programming by expressing what computation desired without specifying how it carries it out.
Often a series of assertions and queries.
30
Algorithm
A series of steps used to solve a particular problem
or
Precise sequences of instructions for processes that an e executed by a computer and are implemented using precise programming languages
31
What is the Luhn algorithm?
Algorithm for credit card verification number.
32
what is deflate?
Algorithm which provides lossless data compression.
33
What are the building blocks of algorithms
sequencing (do this then do that)
selection (use booleans to select which parts to do)
iteration (repeat until)
recursion
34
Properties of algorithm
- algs can be combined
- knowledge of standard algs can help build new algs
- ensuring old lags are correct is necessary for creating new algs
- different lags can be developed for same problem
- developing new lag to solve a problem can yield insight into problems
35
Language for algorithms
- natural languages, pseudo code
| - visual and text based programing languages
36
What is the difference between Algorithms and functions
```
Algs= conceptual definitions of who to accomplish a task. they are language agnostic
Functions= implementation of algorithms in a particular language
```
37
Programming languages
```
C/C++ = good for programming that is close to hardware
Java/C# = portable code
Python/Perk/TclTK = fast to write and portable
```
38
Turing Completeness
If a language is turing complete i it can simulate a Universal Turing Machine.
Therefore all languages which are turing complete are equally powerful.
39
When is an algorithm correct?
When every input reports the correct output and doesn't run forever or cause an error.
40
What is a GWAP?
A game with a purpose.
This is is a game that produces some sort of useful or valuable output as a result of playing it. In the case of Google Image Labeler (the ESP game), playing the game helps Google identify what the subject of a picture is.
41
Who is Luis von Ahn?
Leader of the GWAP movement
42
Koans
= paradoxes
| They describe a world that is developing but hasn’t yet fully emerged
43
What are the 7 Koans?
=paradoxes
1. It's all just bits
2. Perfection is normal
3. There is want in the midst of plenty
4. Processing is power
5. More of the same
6. Nothing goes away
7. Bits move faster than thought
44
Jack Kilby
Inventor of the first integrated circuit (1957)
45
John Bardeen and Walter Brattain
Inventor of the first transistor (half an inch tall) in 1947
46
planar transistor
components are etched directly onto a semiconductor substrate. then metal film applied which adheres the raised parts of the semiconductor materials
47
Wh invented planar transistors?
Jean Hoerni developed the technique for creating planar transistors in 1959.
48
Seven steps of search engines
1) Gather information
2) Keep copies
3) Build an Index
4) Understand the query
5) Relevance of each possible result
6) Ranking of best relevant results
7) Present results
49
Koomey's law
Amount of energy required by fixed computing load drops every 18 months
50
Deadlock
Two programs/thread wait for other to finish
51
Race condition
two programs/threads try to access the same piece of data at the same time
52
Amdhal's law
Applications can almost never be completely parallelized, some serial code remains because you require something to be done first.
53
Deep mind
Company that trained machines to play Atari games
54
John McCarthy's definition of AI
"Getting a computer to do things which when done by humans are said to involve intelligence."
55
Different components of AI
- machine learning
- formal logic
- big data
56
Iterative design cycle
Evaluate-->Design--> prototype--->
57
How to understand users to improve HCI
- choosing existing practices
- create scenarios of actual use (capture commonalities of actual use)
- build models to gain insight
58
What is HCI?
human computer interaction
59
How much code is devoted to user interface (UI) ?
48%
60
Different eras of computing
Mainframe computing
Personal Computing
Ubiquitous
61
Brian Harvey on Copyright
We're going to make a bargain with creators. We're going to give you a limited time monopoly to profit form your idea in return for sharing your idea with us"
62
DMCA
Digital Millenium Copyright Act
63
Tom Mitchell
A machine learns if after an experience it performs better
64
Algortihm types for machine learning
- Supervised learning
- Unsupervised learning
- reinforced learning
65
What was TCP?
Invented by Robert Kahn and Vint Cerf in 1973 = now part of the Internet Protocol Suite
66
DNS
Domain Name System
| Translates names to IP addresses
67
IP adress
Enables connecting new devices to internet.
| IPV6 using Ox now being introduced
68
IETF
Internet Engineering Task Force
| Develop and oversee standrads such as HTTP, IP, SMTP
69
Mosaic
1st popular WWW browser
70
Packets
Blocks of bits data is sent in across internet
71
Bandwith
A measure of bit rate—the amount of data (measured in bits b) that can be sent in a fixed time. Usually b/s
72
Latency
The time elapsed between a transmission and receipt of request. Usually measured in ms
73
Symmetric encryption
A method of encryption involving one key for encryption and decryption
74
Public key encryption
=not symmetric
| =an encryption method that is widely used because of the functionality it provides
75
Lossless compression
Entropy compression reduces data by removing redundant information.
76
Lossy compression
Eg MP3.
Removes irrelevant information
Not fully reversible
