kudos for citing Searn, it's one of the greatest joint prediction and learning algorithms that started a whole range of useful learning to search approaches.<p>LOLS is even better. all approaches inspired by the reinforcement learning but use the information from already present data to imitate the constructed policy.<p>training neural networks with the approach of LOLS could make so many tasks reachable.<p>one could probably even imitate a komodo 9 algorithm with it.