M2.2 Flashcards

1
Q

Three possible practical implementations of ANNs are:
1. A software simulation program running on a digital computer.
2. A hardware emulator connected to a host computer - the so-called _____.
3. A true electronic neural network

A

neurocomputer

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

This is currently the cheapest and fastest implementation method for ANNs

A

Software Simulations

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Software Simulations simulates parallel processing on a conventional ___ digital computer

A

sequential

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Software Simulations replicates ___behaviour of the network by updating the activation level and output of each node for successive time steps

A

temporal

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Typical additional features of ANN simulators
1. Configuring the net according to a chosen architecture and node operational characteristic
2. Implementation of ___ phase using a chosen training algorithm

A

training

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

ANN simulators are written in ___ languages such as C, C++ and Java

A

hi-level

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Main attraction of ANN simulators is the relatively low cost and wide availability of ready made commercial ___

A

packages

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

ANN simulators are also compact, flexible and highly ___

A

portable

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Training of ANNs using software simulators can be slow for larger networks (greater than a few ___)

A

hundred

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Commercially Available Neural Net Packages have prewritten ___ with convenient user interfaces

A

shells

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Commercially Available Neural Net Package cost a few ___ of dollars

A

hundred to tens of thousands

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Commercially Available Neural Net Package allow users to specify the ANN design and training ____

A

parameters

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Commercially Available Neural Net Package usually provide ___ to enable monitoring of the net’s training and operation

A

graphic interfaces

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Commercially Available Neural Net Package are likely to provide interfacing with other software systems such as ___ and databases

A

spreadsheets

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

___ are dedicated special-purpose digital computer (aka accelerator boards

A

neurocomputer

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Neurocomputers are dedicated special-purpose digital computer (aka ___)

A

accelerator boards

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

Neurocomputers acts as a ___to a host computer and is controlled by a program running on the host

A

coprocessor

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

Neurocomputers can be tens to ___ of times faster than simulators

A

thousands

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

Neurocomputers systems are available with over 10 million ___ for networks with several million neurons

A

IPS

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

In neurocomputers, IPS means?

A

IPS (Interconnect updates Per Second)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

___ in hardware are closer to biological neural networks

A

True Networks

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

True Networks in Hardware are consist of synthetic neurons actually fabricated on ___ chips

A

silicon

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

Commercially available ___ ANNs are limited to a few thousand neurons per chip

A

hardwired

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

The problem with true networks in hardware wherein it is hard to make all the neurons communicate properly across chips

A

Interconnection and interference issues

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
Q

The problem with true networks in hardware wherein some chips can’t change their “memory strength” after being made

A

fixed-valued weights

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
26
Q

The problem with true networks in hardware wherein scientists are still working on making the connections changeable or a modifiable ___

A

synapses

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
27
Q

This aims to add structure and organization to ANN applications development for reducing cost, increasing accuracy, consistency, user confidence and friendliness

A

Neural Network Development Methodology

28
Q

Neural Network Development Methodology are split into (4) phases:

A

The Concept Phase
The Design Phase
The Implementation phase
The Maintenance Phase

29
Q

In Neural Network Development Methodolog, this phase involves
o Validating the proposed application
o Selecting an appropriate neural paradigm.

A

The Concept Phase

30
Q

One application area unsuitable for ANNs is ___ management eg, inventory, accounts, sales data analysis

31
Q

Selecting an ANN Paradigm is based on comparison of application requirements to capabilities of different paradigms as well as the ___ method that can be employed

32
Q

This phase of Neural Network Development Methodology specifies initial values and conditions at the node, network and training levels

A

The Design Phase

33
Q

In Node-Level Decision of the the Design Phase, this type of input is like a light switch: ON or OFF (0,1)

34
Q

In Node-Level Decision of the the Design Phase, is like YES/NO with more balance (-1, +1)

35
Q

In Node-Level Decision of the the Design Phase, this type of input has three options: low, neutral, high (-1, 0, +1)

36
Q

In Node-Level Decision of the the Design Phase, these are the other two types of input (2)

A

Discrete, Continuous

37
Q

In Node-Level Decision of the the Design Phase, what is this type of Transfer Function

If input is big enough, turn ON; else, stay OFF

A

Step/Threshold

38
Q

In Node-Level Decision of the the Design Phase, what is this type of Transfer Function

Smooth curve output between 0 and 1

39
Q

In Node-Level Decision of the the Design Phase, what is this type of Transfer Function

Smooth curve from -1 to +1

A

Tang (hyperbolic tangent)

40
Q

In Node-Level Decision of the the Design Phase, what is this type of Transfer Function

Precomputed answers to speed things up

A

Lookup Table

41
Q

In Network-Level Decision of the the Design Phase, __ is described as how many inputs you give (like pixels of an image)

A

Input layer size

42
Q

In Network-Level Decision of the the Design Phase, __ is described as how many categories or clusters you expect

A

Output layer size

43
Q

In the Design Phase, this type of connectivity is when neurons only connect between layers (like in Multilayer Perceptron or MLP)

A

Interlayer Connectivity

44
Q

In the Design Phase, this type of connectivity is when neurons also connect within the same layer (like in Hopfield networks)

A

Intralayer Connectivity

45
Q

In the Design Phase, __ feedback is once the signal moves forward, it doesn’t go back (like in MLP)

A

no feedback

46
Q

In the Design Phase, __ feedback is when outputs can be sent back into the system to help adjust behavior (like in Hopfield net)

A

with feedback

47
Q

This parameter in the Design Phase is defined as how fast the network updates its weights

A

learning rate

48
Q

This parameter in the Design Phase is defined as when should training stop? (e.g., after 1000 rounds, or when error is small enough)

A

stopping rule

49
Q

This parameter in the Design Phase is defined as adding small random numbers cto help the network learn faster (like shaking things up to avoid getting stuck)

A

adding noise

50
Q

What is the last step in the implementation step of Neural Network Development Methodology

o Gathering the training set
o Selecting the developing environment
o Implementing the neural network
o ___

A

Testing and debugging the network

51
Q

In gathering training data in the implementation step, Increasing data amount increases ___ time but may help earlier convergence

52
Q

In gathering training data in the implementation step, quality more important than quantity. True or false

53
Q

In the implementation step, when preparing for data, it usually includesnormalization and possible ___

A

binarization

54
Q

In the implementation step, selecting the __ environment is also included. Such as picking the right hardware and software aspects

A

Development Environment

55
Q

In the implementation phase, the most popular platforms to consider as the development environment are workstations and high-end PC’s (with ___ board option)

A

accelerator

56
Q

In the implementation phase, when choosing the software, this type of software requires more expertise on part of the user but provides maximum flexibility

A

Custom-coded simulators

57
Q

In the implementation phase, when choosing the software, this type of software are usually easy to use because of a more sophisticated interface

A

Commercial developement packages

58
Q

This phase of Neural Network Development Methodology is consists of
o placing the neural network in an operational environment with possible integration
o periodic performance evaluation, and maintenance.

A

The Maintenance Phase

59
Q

In the Maintenance Phase, when putting the Network to Work, this type of setup Works by itself, like a calculator

A

Stand-alone

60
Q

In the Maintenance Phase, when putting the Network to Work, this type of setup Works with other systems across a network

A

Distributed

61
Q

In the Maintenance Phase, when putting the Network to Work, this type of setup Prepares data for another system

A

Preprocessor

62
Q

In the Maintenance Phase, when putting the Network to Work, this type of setup Processes the output from another system

A

Postprocessor

63
Q

In the Maintenance Phase, when putting the Network to Work, this type of setup Built directly into another system (like a smart fridge or robot)

64
Q

In the Maintenance Phase, when putting the Network to Work, this type of connection to other systems is when ANN works alongside other systems but is not deeply connected. Examples: preprocessor, postprocessor, separate module

A

Loose-coupling

65
Q

In the Maintenance Phase, when putting the Network to Work, this type of connection to other systems is when ANN is fully integrated into another system — it feels like part of the main program or machine

A

Tight-coupling

66
Q

In the Maintenance Phase, the two main wais to AN maintenance is (2)

A

Data Maintenance
Software Maintenance