Transfer learning: the ultimate cop-out of deep learning. "I've got a model, but I don't have to actually train it myself, because I'm lazy."
But seriously, folks, transfer learning is a thing, and it's a game-changer. You can take an existing model, slap some new data on it, and voila! You're a deep learning rockstar.
But don't just take our word for it. Here are some examples: