I am looking for some deep learning applied to study of trajectories. I have looked on the internet but it looks like very little has been done in this field. Does anyone have some suggestions?
By trajectories, do you mean something similar to location based Point of Interest guidance systems?
There are search results for this : guessing the next Point-of-Interest, learning from previous POI travel patterns etc.
For example this paperA Survey of Point-of-Interest Recommendation
in Location-Based Social Networks
I’ve attempted to use an LSTM for this before, I guess most of the people use Markov models/probabilistic suffix trees. Would be interesting to try!
If that’s what the question means, then definitely check out lesson 14 of the last course - there’s a taxi destination prediction competition that uses embeddings to win.