Artificial Neural Networks

Mission

My mission is to make machines intelligent and improve people’s lives. We must create machines that have a better understanding of data at the same level as humans.

HOW?

With artificial neural networks, what is artificial neural networks?

Artificial neural networks (ANNs) or connectionist systems is a system inspired by the biological neural networks found in human or animals. This system learns tasks by considering examples, generally without task-specific programming. If we give this kind of system an image it might learn to identify images that contain dogs by analyzing example images that have been manually labeled as "dogs" or "no dogs" and using the results to identify dogs in other images. They do this without any a priori knowledge about dog, but it uses examples like they have fur, they have long ears, dog like faces. These systems evolve their own set of relevant characteristics from the learning material that they have gathered and processed.
Artificial neural networks (ANNs) or connectionist systems is a system that has a set of nodes, and connections between nodes. The nodes can be seen as computational units. They receive inputs, and process them to obtain an output. This processing might be very simple or quite complex, the connections determine the information flow between nodes.
Typically, artificial neurons are organized in layers. Different layers may perform different kinds of transformations on their inputs. Signals travel from the first (input), to the last (output) layer, possibly after traversing the layers multiple times.
The original goal of the ANN approach was to solve problems in the same way that a human brain would. However, over time, attention focused on matching specific tasks, leading to deviations from biology. ANNs have been used on a variety of tasks, including computer vision, speech recognition, machine translation, social network filtering, playing board, video games and medical diagnosis.

The history of artificial neural networks

1940’s - D.O. Hebb created a learning hypothesis based on the mechanism of neural plasticity that became known as Hebbian learning. Hebbian learning is unsupervised learning.
1943 - McCulloch and Pitts introduced the first neural network computing model. Their computational model for neural networks was based on mathematics and algorithms this model was split into two sections. One approach focused on biological processes in the brain while the other focused on the application of neural networks to artificial intelligence.
1950's - Rosenblatt's work resulted in a two-layer network, the perceptron, which was capable of learning certain classifications by adjusting connection weights. Although the perceptron was successful in classifying certain patterns, it had a number of limitations. The perceptron was not able to solve the classic XOR (exclusive or) problem. Such limitations led to the decline of the field of neural networks. However, the perceptron had laid foundations for later work in neural computing.
1954- Farley and Clark first used computational machines, then called "calculators", to simulate a Hebbian network. In the late 1980s research expanded to low-level.
Early 1980's -researchers showed renewed interest in neural networks. Recent work includes Boltzmann machines, Hopfield nets, competitive learning models, multilayer networks, and adaptive resonance theory models.

How does an artificial neural network function?

An (artificial) neural network is a network of simple elements called neurons, which receive input, change their internal state (activation) according to that input, and produce output depending on the input and activation. The network forms by connecting the output of certain neurons to the input of other neurons forming a directed, weighted graph. The weights as well as the functions that compute the activation can be modified by a process called learning which is governed by a learning rule.

Figure A

We input a picture the machine will then scan and analyse this picture and try to determine whether it is a cat or dog.

Figure B

To try and get the best answer it will transmit the information into the input layer, then the activated neuron will try to read the image and identify what animal this is. It will send this information into the output layer, which is the final stage that then tell you whether the image is of a dog or cat.

What are artificial neural networks used FOR?

• Calculator
• Medicine, storing medical records based on case information
• Speech production: reading text aloud
• Vision: face recognition , edge detection, visual search engines
• Financial Applications: time series analysis, stock market prediction.
• Data Compression: speech signal, image.
• Game Playing: chess…
• Parking assist cameras in cars.

Advantages and disadvatages

Advantages

• A neural network can perform tasks that a linear program cannot.
• When an element of the neural network fails, it can continue without any problem by their parallel nature.
• A neural network learns and does not need to be reprogrammed.
• It can be implemented in any application.
• It can be implemented without any problem.
• They can work fine in case of incomplete information
• They do not require knowledge of the algorithm solving the problem (automatic learning).
• Process information in a highly parallel way
• They can generalize (generalize to cases unknown)
• They are resistant to partial damage.
• They can perform associative memory (associative - like working memory in humans) as opposed to addressable memory (typical for classical computers)

Disadvantage

• The neural network needs training to operate.
• The architecture of a neural network is different from the architecture of microprocessors therefore needs to be emulated.
• Requires high processing time for large neural networks.
• Black box: It is not possible to explain how the results were calculated in any meaningful way.
• Optimizing parameters: There are many parameters to be set in a neural network and optimizing the network can be challenging, especially to avoid overtraining.

Artificial neural network applications

Deep dream

DeepDream is a computer vision program created by Google engineer Alexander Mordvintsev which uses a convolutional neural network to find and enhance patterns in images via algorithmic pareidolia, thus creating a dream-like hallucinogenic appearance in the deliberately over-processed images.

Speech to Text

There are different approaches to speech recognition like Hidden Markov Model (HMM), Dynamic Time Warping (DTW), Vector Quantization (VQ), etc. This paper provides a comprehensive study of use of Artificial Neural Networks (ANN) in speech recognition.
Start writing here...