4

For a school project, I would like to investigate a paper on either reinforcement learning or computer vision. I am particularly interested in DQN, RNNs, CNNs or LSTMs. I would eventually like to implement any of these. However, I also need to take into account the computing resources required to train and analyse any of these algorithms. I understand that, in computer vision, the data sets can be quite large, but I am not so sure regarding the resources needed to implement and train a typical state-of-the-art RL algorithm (like DQN).

Would a "standard PC" be able to run any of these algorithms decently to achieve some sort of analysis/results?

nbro
  • 42,615
  • 12
  • 119
  • 217
Wizard
  • 313
  • 1
  • 2
  • 6

1 Answers1

1

It might be hard to implement deep reinforcement learning algorithms, especially considering your previous experience and the computing resources you have. They require almost the same (even more) GPU power. Deep reinforcement learning algorithms use deep neural networks for learning the optimal policy. Even if you are given appropriate resources, it would be tough to replicate the results of the paper, if you are a novice.

nbro
  • 42,615
  • 12
  • 119
  • 217
thecomplexitytheorist
  • 776
  • 3
  • 19