The UCB1 algorithm uses the following formula when determining which arm to pull in the Multi-Armed Bandit problem:

Which of the following describes best the role of the parameter y:

Question

The UCB1 algorithm uses the following formula when determining which arm to pull in the Multi-Armed Bandit problem: 
  
 Which of the following describes best the role of the parameter y:

Accepted Answer

y trades off Exploration and Exploitation. Higher values of y increase exploration and decrease exploitation.

Answer

y is the Learning Rate. Higher values of y may result in faster convergence, but allow only smaller updates.

Answer

y is a Regularization parameter. Higher values of y can better deal with noise in the training examples, but may lead to a worse fit on the training data.

Answer

y is a Regularization parameter. Higher values of y may lead to a better fit on the training examples, but may be more susceptible to noise.

Answer

y trades off Exploration and Exploitation. Higher values of y increase exploitation and decrease exploration.

Answer

y is the Learning Rate. Higher values of y allow larger updates, but may result in slower convergence.

Crowdly

The UCB1 algorithm uses the following formula when determining which arm to pull...

Want instant access to all verified answers on moodle.jku.at?