Neural Networks: Types and Applications

DL techniques are classified into three major categories: unsupervised, partially supervised (semi-supervised) and supervised. Furthermore, deep reinforcement learning (DRL), also known as RL, is another type of learning technique, which is mostly considered to fall into the category of partially supervised (and occasionally unsupervised) learning techniques.

Deep supervised learning

This technique deals with labeled data. When considering such a technique, the environs have a collection of inputs and resultant outputs $(x_t,y_t)\sim \rho$. For instance, the smart agent guesses $\hat y = f(x_t)$ if the input is xt and will obtain $l (\hat y _t, y_t)$ as a loss value. Next, the network parameters are repeatedly updated by the agent to obtain an improved estimate for the preferred outputs. Following a positive training outcome, the agent acquires the ability to obtain the right solutions to the queries from the environs. For DL, there are several supervised learning techniques, such as recurrent neural networks (RNNs), convolutional neural networks (CNNs), and deep neural networks (DNNs). In addition, the RNN category includes gated recurrent units (GRUs) and long short-term memory (LSTM) approaches. The main advantage of this technique is the ability to collect data or generate a data output from the prior knowledge. However, the disadvantage of this technique is that decision boundary might be overstrained when training set doesn't own samples that should be in a class. Overall, this technique is simpler than other techniques in the way of learning with high performance.

Deep semi-supervised learning

In this technique, the learning process is based on semi-labeled datasets. Occasionally, generative adversarial networks (GANs) and DRL are employed in the same way as this technique. In addition, RNNs, which include GRUs and LSTMs, are also employed for partially supervised learning. One of the advantages of this technique is to minimize the amount of labeled data needed. On other the hand, One of the disadvantages of this technique is irrelevant input feature present training data could furnish incorrect decisions. Text document classifier is one of the most popular example of an application of semi-supervised learning. Due to difficulty of obtaining a large amount of labeled text documents, semi-supervised learning is ideal for text document classification task.

Deep unsupervised learning

This technique makes it possible to implement the learning process in the absence of available labeled data (i.e. no labels are required). Here, the agent learns the significant features or interior representation required to discover the unidentified structure or relationships in the input data. Techniques of generative networks, dimensionality reduction and clustering are frequently counted within the category of unsupervised learning. Several members of the DL family have performed well on non-linear dimensionality reduction and clustering tasks; these include restricted Boltzmann machines, auto-encoders and GANs as the most recently developed techniques. Moreover, RNNs, which include GRUs and LSTM approaches, have also been employed for unsupervised learning in a wide range of applications. The main disadvantages of unsupervised learning are unable to provide accurate information concerning data sorting and computationally complex. One of the most popular unsupervised learning approaches is clustering.

Deep reinforcement learning

Reinforcement Learning operates on interacting with the environment, while supervised learning operates on provided sample data. This technique was developed in 2013 with Google Deep Mind. Subsequently, many enhanced techniques dependent on reinforcement learning were constructed. For example, if the input environment samples: $x_t\sim \rho$, agent predict: $\hat y _t = x(f_t)$ and the received cost of the agent is $c_t \sim \ho (c_t | x_t, \hat y_t)$ , P here is the unknown probability distribution, then the environment asks a question to the agent. The answer it gives is a noisy score. This method is sometimes referred to as semi-supervised learning. Based on this concept, several supervised and unsupervised techniques were developed. In comparison with traditional supervised techniques, performing this learning is much more difficult, as no straightforward loss function is available in the reinforcement learning technique. In addition, there are two essential differences between supervised learning and reinforcement learning: first, there is no complete access to the function, which requires optimization, meaning that it should be queried via interaction; second, the state being interacted with is founded on an environment, where the input $x_t$ is based on the preceding actions.

For solving a task, the selection of the type of reinforcement learning that needs to be performed is based on the space or the scope of the problem. For example, DRL is the best way for problems involving many parameters to be optimized. By contrast, derivative-free reinforcement learning is a technique that performs well for problems with limited parameters. Some of the applications of reinforcement learning are business strategy planning and robotics for industrial automation. The main drawback of Reinforcement Learning is that parameters may influence the speed of learning. Here are the main motivations for utilizing Reinforcement Learning:

• It assists you to identify which action produces the highest reward over a longer period.

• It assists you to discover which situation requires action.

• It also enables it to figure out the best approach for reaching large rewards.

• Reinforcement Learning also gives the learning agent a reward function.

Reinforcement Learning can't utilize in all the situation such as:

• In case there is sufficient data to resolve the issue with supervised learning techniques.

• Reinforcement Learning is computing-heavy and time-consuming. Specially when the workspace is large.