Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
40 Cards in this Set
- Front
- Back
|
What is Dijkstra's algorithm? |
An algorithm that finds the shortest path between 2 nodes or vertices in a graph
|
|
|
What is node? |
Fundamental unit from which graphs are formed |
|
|
How does Dijkstra's algorithm works? |
1.Give the start vertex a final value of 0
2. Give each vertex connected to the vertex we have just given a final value to a working (temporary) value
3. Check the working value of any vertex that has not yet been assigned a final value.Assign the smallest value to the vertex ; this is now its final value
4. Repeat steps 2 and 3 until the end vertex is reached, and all vertices have been assigned a final value
5. Trace the route back from the end vertex to the start vertex to find the shortestpath between these 2 vertices |
|
|
What is the example of use in Dijkstra's algorithm |
Gps tracking |
|
|
What is an A* Algorithm? |
An algorithm that finds the shortest route between nodes or vertices but uses anadditional heuristic approach to achieve better performance than Dijkstra’salgorithms |
|
|
What is heuristic? |
Heuristic = method that employs a practical solution (rather than a theoreticalone) to a problem; when applied to algorithms this includes running tests andobtaining results by trial and error |
|
|
How does the A* algorithm works? |
It is based on Dijkstra's but adds a heuristic value Step 1. Find h value using Manhattan method Step2.Find g value(movement cost) moving up or down ,left or right .Choose g values based on right angle triangle Step 3. Find F values by using formula: f= g+h |
|
|
What are the examples of applications of the shortest path algorithm include? |
Global positioning satellite (GPS) Google Maps IP routine Modelling the spread of infections diseases |
|
|
What is artificial intelligence? |
Artificial IntelligenceThe branch of computer science that aims to make machines ‘intelligent’ A machine with cognitive abilities such as problem solving and learning from examples |
|
|
What are the 3 categories of Artificial intelligence? |
1) Narrow AI -> A machine that has superior performance to a humans when doing one specific task 2) General AI -> A machine that has same performance to a human when doing any intellectual task 3)Strong AI -> A machine that has superior performance to a human when performing many tasks |
|
|
What are the examples of AI? |
News generation based on live news feed Smart Home devices |
|
|
What is artificial neural networks? |
Neural Networks = an algorithm loosely inspired by the brain, have revolutionized manyfields, including computer vision |
|
|
How artificial neural networks have helped with machine learning |
Can Recognise complex patterns |
|
|
What is machine learning? |
A computer program that improves its performance at certain tasks with experience Systems that learns without needing to be a program to learn The algorithm learns from past experiences and examples The system makes a decision or makes predictions based on previous situations They have the ability to manage and analyse large volume of complex data |
|
|
What is deep learning? |
machines that think in a way similar to the human brain
they handle huge amounts of data using artificial neural networks by looking at binary pixels of each pixel They are excellent at identifying patterns which would be to complex or time consuming for humans to carry out It forms algorithms in layers to create an artificial neural networks that can learn and make intelligent decisions on its own It is artificial neural networks are based on the interconnections between neurons in the brain The hidden layers are where data from the input layer is processed into something which can be sent to the output layer Uses artificial neural networks that contains a high number of hidden layers modeled on the human brain Deep learning is a specialized form of machine learning.
|
|
|
What is the example of deep learning |
Face recognition |
|
|
What is labelled data? |
Data where we know the target answer and the object is fully recognised |
|
|
What is unlabelled data? |
Data where the objects are undefined and need to be manually recognised |
|
|
What is supervised learning? |
Using known tasks with given outcomes to enable a computer program to improveits performance in accomplishing similar tasks “Labeling” or predicting unknown value for some piece of data It uses regression analysis and classification analysis The system requires both an input and output to be given to the model so it can be trained The model uses labelled data so the desired output for a given input is known
Algorithms receives a set of inputs and the correct outputs to permit the learning process Once trained, the model is run using labelled data The results are compared with the expected output if there are any errors than the model needs further refinement The model is run with the unlabelled data to predict the outcome |
|
|
What is supervised learning? |
Using known tasks with given outcomes to enable a computer program to improveits performance in accomplishing similar tasks Supervised learning allows data to be collected from the previous experience Able to predict future outcomes based on past data “Labeling” or predicting unknown value for some piece of data It uses regression analysis and classification analysis The system requires both an input and output to be given to the model so it can be trained The model uses labeled data so the desired output for a given input is known
Algorithms receive a set of inputs and the correct outputs to permit the learning process Once trained, the model is run using labelled data The results are compared with the expected output if there are any errors than the model needs further refinement The model is run with the unlabelled data to predict the outcome |
|
|
What is the example of supervised learning? |
Categorising emails as relevant or spam without human intervention |
|
|
What is unsupervised learning? |
The system which is able to identify the hidden patterns from input data
They are not trained using the right answer uses unlabelled input data.
The algorithms evaluate the data to find any hidden patterns or structures within data set Using a large number of tasks with unknown outcomes and the use of feedback to enable a computer program to improve its performance in accomplishing similar tasks Discovering “structure” or underlying patterns in a collection of data E.g. Discovering diseases, finding groups of customers for marketing, exploringdata before you do something else with it |
|
|
What is the example of unsupervised learning? |
Product marketing -> Group of people with the same behaviour are considered a single unit |
|
|
What is reinforcement learning? |
System which is given no training which learns on basis of reward and punishment
It help to increase efficiency of the system by making use of optimisation techniques
Using a large number of tasks with unknown outcomes and the use of feedback to enable a computer program to improve its performance in accomplishing similar tasks
Learn to pick an action based on the state of the world, and “rewards” or“punishments” from previous choices |
|
|
What are the examples of uses of reinforcement learning? |
Search engines Online games Robotics |
|
|
How does an artificial neural network work? |
Systems are able to recognise objects which they form labelled data which can be used in the training process |
|
|
How does deep learning works? |
1) large amounts of unlabelled data goes into the model 2) It recognises objects by looking at the binary codes of each pixel which creates a picture of the object 3) The model is trained using artificial neural networks to identify the objects 4) New data 5) Labelled data goes into the model to make sure it gives the correct responses 6)The required output is provided if the output is not sufficiently accurate the module will be cleared until it gives the good results |
|
|
What is back propagation? |
A method used in artificial neural networks to calculate error gradients so that actual neuron weightings can be adjusted to improve the performance of the model
During development of neural network, weighting must be give to each of the neural connection.The system designer won't know which weight factors produce the best results |
|
|
How does the back propagation works? |
The initial outputs form the system are compared to the expected outputs and thesystem weightings are adjusted to minimise the difference between actual and expected results
Calculus is used to find the error gradient in the obtained outputs: the results arefed back into the neural networks and the weightings on each neuron are adjusted
Once the errors in the output have been eliminated (or reduced to acceptable limits) the neural network is functioning correctly and the model has been successfully set up
If the errors are still too large, the weightings are altered - the process continues until satisfactory outputs are produced |
|
|
How does deep learning enhance the photograph? |
The latest smartphones camera use deep learning to give the DSLR quality to the photographs The technology was developed by taking the same photos using a smartphone and then use a DSLR camera The deep learning system was trained by comparing 2 photographs A large number of photographs already taken by a DSLR camera were used to test the model |
|
|
How does deep learning turn monochrome photos into colour? |
Deep learning can be used to turn monochrome photographs into coloured photographs which produces better photographs than simply turning grey-scale values into an approximated colour Deep learning system is trained by searching websites for data which allows it to recognise features and then map a particular colour to a photograph which produce an accurate coloured image |
|
|
What is the difference between machine learning and deep learning? |
MACHINE LEARNING Enable machines to make decisions on their own based on past data
Need a small amount of data to carry out the training
Most of the features in the data needs to be identified in advance and manually coded in the system
A modular approach is taken to solve a problem
Testing of the system takes a long time to carry out DEEP LEARNING enables machines to make decisions using an artificial neural networks
The system needs large amounts of data during the training stages
Deep learning machine learns the features of the data from the data itself and it does not need to be identified in advance
Testing of the system takes much less time to carry out
|
|
|
What are the 2 types of back propagation? |
1.Static -> Static maps static inputs to a static output ->Mapping is instantaneous -> Training a model is easier than recurrent 2. Recurrent Training a network/model is more difficult Activation is fed forwards until a fixed value is achieved Mapping is not instantaneous |
|
|
What is regression? |
Statistical measure used to make predictions from data by finding learning relationships between the inputs and outputs
It helps understand how the values of a dependent variable changes when the values of independent variables are also changed |
|
|
How does machine learning uses labelled data to recognise the object? |
1) It will see any similarities of the object and recognises it as labelled data which allows it to be trained 2) When it is trained , it would recognise the object from the original data set 3) When these is incoming data, the algorithm analyses it and learn from this data 4) Decisions are made based on what the machine has learned 5)It recognises the new data and produce an output automatically |
|
|
What are the examples of machine learning? |
Spam detection Search engines refining searches based on earlier searches carried out |
|
|
How do graphs aid AI? |
Artificial Neural Networks can be represented using graphs • Graphs provide structures for relationships // graphs providerelationships between nodes • AI problems can be defined/solved as finding a path in a graph • Graphs may be analysed/ingested by a range of algorithms •e.g. A* / Dijksta’s algorithm •used in machine learning. • Example of method e.g. Back propagation of errors / regressionmethods |
|
|
State the reason for having multiple hidden layers in an artificial neural network. |
Enables deep learning to take place |
|
|
Explain how artificial neural networks enable machine learning. |
Artificial neural networks are intended to replicate the way human brains work Weights are assigned for each connection between nodes Backpropagation will be used to correct any errors that have been made. Decisions can be made without being specifically programmed |
|
|
Outline the reasons for using deep learning |
Deep learning outperforms other methods if the data size is large Deep learning is effective at identifying patterns that are too complex or time-consuming for humans to carry out. |
