What is a Dueling Network?

This feature is experimental; we are continuously improving our matching algorithm. A Dueling Network is a type of Q-Network that has two streams to separately estimate (scalar) state-value and the advantages for each action. Both streams share a common convolutional feature learning module.

Does a single clip network perform better than a Dueling Network?

Detailed results are presented in the Appendix. Using this 30 no-ops performance measure, it is clear that the dueling network (Duel Clip) does substantially better than the Single Clip network of similar capacity. It also does considerably better than the baseline (Single) of van Hasselt et al. (2015).

What is a dueling architecture?

(Wang et al.) presents the novel dueling architecture which explicitly separates the representation of state values and state-dependent action advantages via two separate streams. The key motivation behind this architecture is that for some games, it is unnecessary to know the value of each action at every timestep.

What is the difference between a single-stream architecture and a dueling architecture?

The single-stream architecture is a three layer MLP with 50 units on each hidden layer. The dueling ar-chitecture is also composed of three layers. After the first hidden layer of 50 units, however, the network branches off into two streams each of them a two layer MLP with 25 hid-den units. The results of the comparison are summarized in Figure 3.