4 ms·

> ability to time travel for training data generation What now?

• nikhilsimha 2y ago
  Pardon the jargon. But it is a necessary addition to the vocabulary. To evaluate if a feature is valuable, you could attach the value of the feature to past inferences and retrain a new model to check for improvement in performance. But this “attach”-ing needs the feature value to be as of the time of the past inference.
  • mulmen 2y ago
    That’s not a new concept.
    • better365 2y ago
      True. But it is not necessary to reinvent the wheel for engineers. :)
      • mulmen 2y ago
        That’s the point of this subthread though. What’s the new thing Chronon is doing? It can’t just be point in time features because that’s already a thing.