I just learn Machine learning and some ANN for a while and still need to figure it out the big picture of it.
I'm still learning the basic and terminology to deepen my knowledge.
I have learn about Reinforcement learning and what i understand (please correct me if i wrong) there 3 grouping method learning.
unsupervised (example for this is restricted boltzman machine)
supervised (CNN)
reinforcement (EKF, Particle Filter)
When I learn about recurrent net, some said that it belongs to supervised learning.
But when i see how it works, it more suitable to said that it belong to reinforcement learning.
can anyone clarify is recurrent net belong to supervised or reinforcement learning?
RNN is always used in supervised learning, because the core functionality of RNN requires labelled data sent in serially.
Now you must have seen RNN in RL too, but the catch is current deep reinforcement learning use the concept of supervised RNN which acts as a good feature vector for agent inside the RL ecosystem.
In simpler terms, the agent, the reward shaping, the environment everything is RL, but the way the deep network in agent learns is using RNN(or CNN or any type of ANN depending upon the problem statement).
So in short RNN always requires labelled data and hence supervised learning, but it can be used in RL environment too.
Supervised learning vs. reinforcement learning. It is almost the same. In supervised learning there is a finite amount of labelled examples. Each example is self standing. All the examples come from the same distribution. If the example is a series of inputs (ex. a sentence made out of words), it is still a single example (ex. "What a lovely day" -> positive).
With RL there are no examples and and at the same time there are infinite number of examples. What??!? Yes, your agent can interact with an environment and this will generate a lot of episodes (ex. "Start -> Left -> 1, 2 -> Up -> 2, 4.."). Or also a sentence (ex. "What -> ah -> a -> go on -> lovely -> not again -> day".). And what is the label? Not clear. Some reward mechanism should be developed to communicate the desired behavior. Also note that the episodes are dependent on the actions the agent takes. So no more "same distribution".
Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 25 days ago.
Improve this question
What's the difference between reinforcement learning, deep learning, and deep reinforcement learning? Where does Q-learning fit in?
Reinforcement learning is about teaching an agent to navigate an environment using rewards. Q-learning is one of the primary reinforcement learning methods.
Deep learning uses neural networks to achieve a certain goal, such as recognizing letters and words from images.
Deep reinforcement learning is a combination of the two, using Q-learning as a base. But instead of using actual state-value pairs, this is often used in environments where the state-action space is so large that it would take too long for Q-learning to converge. By using neural networks, we can find other state-action pairs that are similar. This “function approximation” allows effective learning in environments with very large state-action spaces.
Deep learning is a method using neural networks to make function approximators to solve various problems.
Ex: Learning a function which takes an image as input and output the bounding boxes of objects in the image.
Reinforcement learning is a field in which we have an agent and we want that agent to perform a task i.e, goal based problems where we use trial and error learning methods.
Ex: Agent learning to move from one position on grid world to a goal position without falling in a pit present in between.
Deep reinforcement learning is a way to solve goal based problems using neural networks. This is because, when we want agents to perform task in real world or current games, the state space is very big.
It takes agent very long time to even visit each state once and we cannot use look up tables to store the value functions.
So, to tackle this problem we use neural networks to approximate the state to generalize the learning process
Ex: We use DQN to solve many atari games.
Q-learning : It is a temporal difference learning method, where we have a Q-table to look for best action possible in the current state based on Q value function.
For learning Q values we use the reward and the maximum possible next state Q value.
Q-learning basically falls under Reinforcement learning and its deep reinforcement learning analog is Deep Q network (DQN).
The goal of machine learning methods is to learn rules from data and make predictions and/or decisions based on them.
The learning process can be done in a(n) supervised, semi-supervised, unsupervised, reinforcement learning fashion.
In reinforcement learning (RL), an agent interacts with an environment and learns an optimal policy, by trial and error (using reward points for successful actions and penalties for errors). It is used in sequential decision making problems [1].
Deep learning as a sub-field of machine learning is a mathematical framework for learning latent rules in the data or new representations of the data at hand. The term "deep" refer to the number of learning layers in the framework. Deep learning can be used with any of aforementioned learning strategies, i.e., supervised, semi-supervised, unsupervised, and reinforcement learning.
A deep reinforcement learning technique is obtained when deep learning is utilized by any of the components of reinforcement learning [1]. Note that Q-learning is a component of RL used to tell an agent that what action needs to be taken in what situation. Detailed information can be found in [1].
[1] Li, Yuxi. "Deep reinforcement learning: An overview." arXiv preprint arXiv:1701.07274 (2017).
Reinforcement learning refers to finish -oriented algorithms, which learn how to attain a coordination compound objective (goal) or maximize along a particular dimension over many steps. The basic theme behind Reinforcement learning is that an agentive role will learn from the environment by interacting with it and getting rewards for performing actions.
Deep Learning uses multiple layers of nonlinear processing units to extract feature and transformation
Deep Reinforcement Learning approach introduces deep neural networks to solve Reinforcement Learning problems thus they are named “deep.”
There's more distinction between reinforcement learning and supervised learning, both of which can use deep neural networks aka deep learning. In supervised learning - training set is labeled by a human (e.g. AlphaGo). In reinforcement learning (e.g. AlphaZero)- the algorithm is self-taught.
To put it in simple words,
Deep Learning - It's uses the model of neural network(mimicking the brain , neurons) and deep learning is used in image classification , data analyzing and in reinforcement learning too.
Reinforcement learning - This is a branch of machine learning, that revolves around an agent (ex: clearing robot) taking actions(ex: moving around searching trash) in it's environment(ex:home) and getting rewards(ex: collecting trash)
Deep-Reinforcement learning - This is one among the list of algorithms reinforcement learning has , this algorithm utilizes deep learning concepts.
Reinforcement learning (RL) is a type of machine learning that is mainly motivated by the feedback control of systems. RL is usually considered a type of optimal control that learns through interacting with a system/environment and getting feedback. RL usually replaces the computationally expensive dynamic programming methods with single time-step/multi time-step learning rule. Popular temporal difference methods in RL are considered somewhere in between dynamic programming and monte carlo methods. Classic RL methods use tabular algorithms that are not that scalable.
Deep learning (DL) is considered crucial part of modern machine learning (classical machine learning usually mean SVM, liner regression etc.). DL uses deep multilayered neural networks (NN) with backpropagation for learning. By using well designed deep NN networks complex input-output relations can be learned. Because of this property of approximating very complex functions DL have been extremely popular in recent years (2010-ish), especially in natural language tasks and computer vision tasks. One of the attractive aspect of DL is that these models can be end-to-end, meaning we do not need to do manual feature engineering. There are numerous types of DL algorithms, like Deep neural networs, convolutional neural networks, GRU, LSTM, GAN, attention, transfromer etc.
Deep RL uses deep NN architectures to replace the tabular methods for very high dimensional problems. Informally speaking, the controller is no longer a table look-up rather we use a deep NN as the controller. Because of leveraging deep NN in RL this is commonly known as deep RL.
roughly speaking:
deep learning uses deep neural networks to approximate complicated functions.
reinforcement learning is a branch in machine learning where your learner learns through interaction with environment. It is different from supervised or unsupervised learning.
if you use deep learning to approximate functions in reinforcement learning you call it deep reinforcement learning.
Reinforcement learning is a type of artificial intelligence that aims to model human-like decision-making. It's based on the idea that humans learn from their actions and reward themselves for doing things that are good, and punish themselves for doing things that are bad. Reinforcement learning algorithms try to replicate this process by changing the value of some variable in response to an action.
Deep learning is a type of machine learning model which uses multiple layers of processing to solve problems more effectively than traditional approaches. Deep learning models can be used for image recognition, speech recognition, and translation.
Deep reinforcement learning is a type of deep learning model that tries to solve problems by using sequences of actions called episodes to improve over time as well as by comparing results from different episodes. It's also known as Q-learning because it was first described by Richard Sutton in 1997 using the Q function (the fourth derivative).
Q-learning is a particular type of deep reinforcement learning algorithm that makes use of Q values (quantified measures) instead of actual rewards or penalties, which means it can be used without having access to real data or rewards/penalties yet still produce useful results
Imagine a binary classification problem like sentiment analysis. Since we have the labels, cant we use the gap between actual - predicted as reward for RL ?
I wish to try Reinforcement Learning for Classification Problems
Interesting thought! According to my knowledge it can be done.
Imitation Learning - On a high level it is observing sample trajectories performed by the agent in the environment and use it to predict the policy given a particular stat configuration. I prefer Probabilistic Graphical Models for the prediction since I have more interpretability in the model. I have implemented a similar algorithm from the research paper: http://homes.soic.indiana.edu/natarasr/Papers/ijcai11_imitation_learning.pdf
Inverse Reinforcement Learning - Again a similar method developed by Andrew Ng from Stanford to find the reward function from sample trajectories, and the reward function can be used to frame the desirable actions.
http://ai.stanford.edu/~ang/papers/icml00-irl.pdf
Closed. This question is off-topic. It is not currently accepting answers.
Want to improve this question? Update the question so it's on-topic for Stack Overflow.
Closed 10 years ago.
Improve this question
What is machine learning ?
What does machine learning code do ?
When we say that the machine learns, does it modify the code of itself or it modifies history (database) which will contain the experience of code for given set of inputs?
What is a machine learning ?
Essentially, it is a method of teaching computers to make and improve predictions or behaviors based on some data. What is this "data"? Well, that depends entirely on the problem. It could be readings from a robot's sensors as it learns to walk, or the correct output of a program for certain input.
Another way to think about machine learning is that it is "pattern recognition" - the act of teaching a program to react to or recognize patterns.
What does machine learning code do ?
Depends on the type of machine learning you're talking about. Machine learning is a huge field, with hundreds of different algorithms for solving myriad different problems - see Wikipedia for more information; specifically, look under Algorithm Types.
When we say machine learns, does it modify the code of itself or it modifies history (Data Base) which will contain the experience of code for given set of inputs ?
Once again, it depends.
One example of code actually being modified is Genetic Programming, where you essentially evolve a program to complete a task (of course, the program doesn't modify itself - but it does modify another computer program).
Neural networks, on the other hand, modify their parameters automatically in response to prepared stimuli and expected response. This allows them to produce many behaviors (theoretically, they can produce any behavior because they can approximate any function to an arbitrary precision, given enough time).
I should note that your use of the term "database" implies that machine learning algorithms work by "remembering" information, events, or experiences. This is not necessarily (or even often!) the case.
Neural networks, which I already mentioned, only keep the current "state" of the approximation, which is updated as learning occurs. Rather than remembering what happened and how to react to it, neural networks build a sort of "model" of their "world." The model tells them how to react to certain inputs, even if the inputs are something that it has never seen before.
This last ability - the ability to react to inputs that have never been seen before - is one of the core tenets of many machine learning algorithms. Imagine trying to teach a computer driver to navigate highways in traffic. Using your "database" metaphor, you would have to teach the computer exactly what to do in millions of possible situations. An effective machine learning algorithm would (hopefully!) be able to learn similarities between different states and react to them similarly.
The similarities between states can be anything - even things we might think of as "mundane" can really trip up a computer! For example, let's say that the computer driver learned that when a car in front of it slowed down, it had to slow down to. For a human, replacing the car with a motorcycle doesn't change anything - we recognize that the motorcycle is also a vehicle. For a machine learning algorithm, this can actually be surprisingly difficult! A database would have to store information separately about the case where a car is in front and where a motorcycle is in front. A machine learning algorithm, on the other hand, would "learn" from the car example and be able to generalize to the motorcycle example automatically.
Machine learning is a field of computer science, probability theory, and optimization theory which allows complex tasks to be solved for which a logical/procedural approach would not be possible or feasible.
There are several different categories of machine learning, including (but not limited to):
Supervised learning
Reinforcement learning
Supervised Learning
In supervised learning, you have some really complex function (mapping) from inputs to outputs, you have lots of examples of input/output pairs, but you don't know what that complicated function is. A supervised learning algorithm makes it possible, given a large data set of input/output pairs, to predict the output value for some new input value that you may not have seen before. The basic method is that you break the data set down into a training set and a test set. You have some model with an associated error function which you try to minimize over the training set, and then you make sure that your solution works on the test set. Once you have repeated this with different machine learning algorithms and/or parameters until the model performs reasonably well on the test set, then you can attempt to use the result on new inputs. Note that in this case, the program does not change, only the model (data) is changed. Although one could, theoretically, output a different program, but that is not done in practice, as far as I am aware. An example of supervised learning would be the digit recognition system used by the post office, where it maps the pixels to labels in the set 0...9, using a large set of pictures of digits that were labeled by hand as being in 0...9.
Reinforcement Learning
In reinforcement learning, the program is responsible for making decisions, and it periodically receives some sort of award/utility for its actions. However, unlike in the supervised learning case, the results are not immediate; the algorithm could prescribe a large sequence of actions and only receive feedback at the very end. In reinforcement learning, the goal is to build up a good model such that the algorithm will generate the sequence of decisions that lead to the highest long term utility/reward. A good example of reinforcement learning is teaching a robot how to navigate by giving a negative penalty whenever its bump sensor detects that it has bumped into an object. If coded correctly, it is possible for the robot to eventually correlate its range finder sensor data with its bumper sensor data and the directions that sends to the wheels, and ultimately choose a form of navigation that results in it not bumping into objects.
More Info
If you are interested in learning more, I strongly recommend that you read Pattern Recognition and Machine Learning by Christopher M. Bishop or take a machine learning course. You may also be interested in reading, for free, the lecture notes from CIS 520: Machine Learning at Penn.
Machine learning is a scientific discipline that is concerned with the design and development of algorithms that allow computers to evolve behaviors based on empirical data, such as from sensor data or databases. Read more on Wikipedia
Machine learning code records "facts" or approximations in some sort of storage, and with the algorithms calculates different probabilities.
The code itself will not be modified when a machine learns, only the database of what "it knows".
Machine learning is a methodology to create a model based on sample data and use the model to make a prediction or strategy. It belongs to artificial intelligence.
Machine learning is simply a generic term to define a variety of learning algorithms that produce a quasi learning from examples (unlabeled/labeled). The actual accuracy/error is entirely determined by the quality of training/test data you provide to your learning algorithm. This can be measured using a convergence rate. The reason you provide examples is because you want the learning algorithm of your choice to be able to informatively by guidance make generalization. The algorithms can be classed into two main areas supervised learning(classification) and unsupervised learning(clustering) techniques. It is extremely important that you make an informed decision on how you plan on separating your training and test data sets as well as the quality that you provide to your learning algorithm. When you providing data sets you want to also be aware of things like over fitting and maintaining a sense of healthy bias in your examples. The algorithm then basically learns wrote to wrote on the basis of generalization it achieves from the data you have provided to it both for training and then for testing in process you try to get your learning algorithm to produce new examples on basis of your targeted training. In clustering there is very little informative guidance the algorithm basically tries to produce through measures of patterns between data to build related sets of clusters e.g kmeans/knearest neighbor.
some good books:
Introduction to ML (Nilsson/Stanford),
Gaussian Process for ML,
Introduction to ML (Alpaydin),
Information Theory Inference and Learning Algorithms (very useful book),
Machine Learning (Mitchell),
Pattern Recognition and Machine Learning (standard ML course book at Edinburgh and various Unis but relatively a heavy reading with math),
Data Mining and Practical Machine Learning with Weka (work through the theory using weka and practice in Java)
Reinforcement Learning there is a free book online you can read:
http://www.cs.ualberta.ca/~sutton/book/ebook/the-book.html
IR, IE, Recommenders, and Text/Data/Web Mining in general use alot of Machine Learning principles. You can even apply Metaheuristic/Global Optimization Techniques here to further automate your learning processes. e.g apply an evolutionary technique like GA (genetic algorithm) to optimize your neural network based approach (which may use some learning algorithm). You can approach it purely in form of a probablistic machine learning approach for example bayesian learning. Most of these algorithms all have a very heavy use of statistics. Concepts of convergence and generalization are important to many of these learning algorithms.
Machine learning is the study in computing science of making algorithms that are able to classify information they haven't seen before, by learning patterns from training on similar information. There are all sorts of kinds of "learners" in this sense. Neural networks, Bayesian networks, decision trees, k-clustering algorithms, hidden markov models and support vector machines are examples.
Based on the learner, they each learn in different ways. Some learners produce human-understandable frameworks (e.g. decision trees), and some are generally inscrutable (e.g. neural networks).
Learners are all essentially data-driven, meaning they save their state as data to be reused later. They aren't self-modifying as such, at least in general.
I think one of the coolest definitions of machine learning that I've read is from this book by Tom Mitchell. Easy to remember and intuitive.
A computer program is said to learn from experience E with respect to some class of tasks T and performance measure P, if its performance at tasks in T, as measured by P, improves with experience E
Shamelessly ripped from Wikipedia: Machine learning is a scientific discipline that is concerned with the design and development of algorithms that allow computers to evolve behaviors based on empirical data, such as from sensor data or databases.
Quite simply, machine learning code accomplishes a machine learning task. That can be a number of things from interpreting sensor data to a genetic algorithm.
I would say it depends. No, modifying code is not normal, but is not outside the realm of possibility. I would also not say that machine learning always modifies a history. Sometimes we have no history to build off of. Sometime we simply want to react to the environment, but not actually learn from our past experiences.
Basically, machine learning is a very wide-open discipline that contains many methods and algorithms that make it impossible for there to be 1 answer to your 3rd question.
Machine learning is a term that is taken from the real world of a person, and applied on something that can't actually learn - a machine.
To add to the other answers - machine learning will not (usually) change the code, but it might change it's execution path and decision based on previous data or new gathered data and hence the "learning" effect.
there are many ways to "teach" a machine - you give weights to many parameter of an algorithm, and then have the machine solve it for many cases, each time you give her a feedback about the answer and the machine adjusts the weights according to how close the machine answer was to your answer or according to the score you gave it's answer, or according to some results test algorithm.
This is one way of learning and there are many more...