Evolving lane merge traffic behaviour simulations via a macroscopic objective function and a machine learning system trained through bootstrapped human judgement.
Erandi LakshikaMichael BarlowAdam EastonPublished in: Appl. Intell. (2016)
Keyphrases
- machine learning
- objective function
- traffic flow
- real time
- agent based simulations
- computer vision
- machine learning methods
- optimization problems
- linear programming
- human decision making
- cognitive model
- decision trees
- human learning
- traffic congestion
- neural network
- human behaviour
- human operators
- learning tasks
- network traffic
- optimal solution
- cost function
- natural language processing
- simulation model
- training set
- data mining
- active learning
- support vector machine
- learning algorithm
- information extraction
- artificial intelligence
- knowledge acquisition
- feasible solution
- lower bound
- cellular automata
- svm classifier
- machine learning algorithms