…Seems so similar to me… I am new in advanced ML techniques, please give me some example.
Solved – the difference between transfer learning and reinforcement learning
reinforcement learningtransfer learning
Related Solutions
Single Task Learning: Giving a set of learning tasks, t1 , t2 , …, t(n), learn each task independently. This is the most commonly used machine learning paradigm in practice.
Multitask Learning: Giving a set of learning tasks, t1 , t2 , …, t(n), co-learn all tasks simultaneously. In other words, the learner optimizes the learning/performance across all of the n tasks through some shared knowledge. This may also be called batch multitask learning. Online multitask learning is more like lifelong learning (see below).
Transfer Learning (or Domain Adaptation): Giving a set of source domains/tasks t1, t2, …, t(n-1) and the target domain/task t(n), the goal is to learn well for t(n) by transferring some shared knowledge from t1, t2, …, t(n-1) to t(n). Although this definition is quite general, almost the entire literature on transfer learning is about supervised transfer learning and the number of source domains is only one (i.e., n=2). It also assumes that there are labeled training data for the source domain and few or no labeled examples in the target domain/task, but there are a large amount of unlabeled data in t(n). Note that the goal of transfer learning is to learn well only for the target task. Learning of the source task(s) is irrelevant.
Lifelong Learning: The learner has performed learning on a sequence of tasks, from t1 to t(n-1). When faced with the nth task, it uses the relevant knowledge gained in the past n-1 tasks to help learning for the nth task. Based on this definition, lifelong learning is similar to the general transfer learning involving multiple source domains or tasks. However, some researchers have a narrower definition of lifelong learning. They regard it as the learning process that aims to learn well on the future task t(n) without seeing any future task data so far. This means that the system should generate some prior knowledge from the past observed tasks to help new/future task learning without observing any information from the future task t(n). The future task learning simply uses the knowledge. This definition makes lifelong learning different from both transfer learning and multitask learning. It is different from transfer learning because transfer learning identifies prior knowledge using the target/future task labeled and unlabeled data. It is different from multitask learning because lifelong learning does not jointly optimize the learning of the other tasks, which multitask learning does.
All content lifted from https://www.cs.uic.edu/~liub/IJCAI15-tutorial.html.
I'd recommend getting an overview of the math that's currently used in deep learning architectures that are used for supervised settings (this does mean looking into approaches that involve "training sets"), before you dive deeper into other math.
http://www.deeplearningbook.org/ has a very good overview of the math you'd need to understand what's going on in neural nets/deep nets. Once you're comfortable with the current approaches, you'd be able to understand the research in the field, and the directions it's heading in. (from ICML, NIPS papers, for instance)
At that point, you will likely find open problems that seem to interest you, and you can begin to actively work on them. It's often useful to have a problem you want to solve in mind, and then explore all the work that's been done on that problem (prior approaches, the math involved, etc) - Sometimes, you'll find that there are some problems that interest you deeply, but the current approaches to solve them are unsatisfying - this is really the point when you might have to invent(discover?) the math needed to solve it, or "borrow" the math from a different field. The main benefit you'll have if you work on problems that are similar to what other researchers are interested in is that there's a community that's publishing work at a breakneck pace, and you'll be able to quickly get feedback on approaches that have been tried and haven't quite worked just yet.
I'm not quite discounting the value of learning math by itself, but just saying that if you learn the math in the light of a problem (or ten), you'll learn how to apply the existing math well (+ how to do a good literature search), and you'll also learn to figure out when new math is required.
Best Answer
Reinforcement learning endeavors to make self-teaching agents which can solve some problem. An example is Google's AlphaGo and AlphaGoZero agents which can teach themselves how to play Chess, Go and Shogi better than any human. The trick here is that in reinforcement learning, the goal is to maximize some reward. We don't tell the agent what the optimal solution is; in the case of chess and Go, it's usually too hard to say which play is the optimal play. So we don't even know how to maximize the reward; figuring out an approximation to the best play is left up to the agent.
This is in contrast to supervised learning contexts where we know the right answer ahead of time, and the model has to learn to match the right answer.
Transfer learning is usually framed as a special case of supervised learning. For example, we might have an image classifier trained on millions of examples for a specific set of classes, but we want to apply it to a different set of classes where data is much more scarce. It can be more effective to only partially retrain the model instead of retraining the whole thing (because the new data is too scarce to effectively train the image classifier). This is still a supervised learning problem, because we're giving labels to the model and telling it to match the labels, but it's different from the standard supervised learning problem because we're adapting the model to the new domain (the different set of classes).
This hot question on AI.SE seems related; https://ai.stackexchange.com/questions/16741/how-can-an-ai-train-itself-if-no-one-is-telling-it-if-its-answer-is-correct-or-w