Are most deep learning models online learning models?

Question

I'm online learning starter. from my perspective, online learning model is the model which can update its paramater with data flows(I've seen a article pointing out that incremental model is irrevalent of time while online learning emphasizes the data flows in time-series).

Here I regard them as one thing.

And in my view, most deep learning can be fine tuned,as we fine-tune a pre-trained bert model, is that means a deep learning model can be fine tuned is equivalent to the deep learning model is a online learning model only if I receive the data flow as soon as possible.

BTW, I Googled online learning, result relevant to this are rarely found, so as to Google scholar. Is this field is not hot anymore? Or I entered wrong keyword,Please let me know. Thank you.

Answer 1

Online learning is defined as a method of machine learning in which data becomes available in a sequential order and is used to update the best predictor for future data at each step, as opposed to batch learning techniques which generate the best predictor by learning on the entire training data set at once, so the time component is important and is the thing which distinguishes between online and offline learning. So I would say "No", most deep learning models are probably not online, but they do use batches when learning, which is a different thing, since the batches are random samples from the data irrelevant of time of data acquisition.