Deep Learning Vs Machine Learning

Human-computer interactions are based on the belief of trust. In a world in which machines are becoming more and more autonomous and capable, safety becomes a major concern. The more artificial intelligence comes into context, the more we need to find a way to measure its trustworthiness.
Additionally, in order to measure the trustworthiness of machines many problems require additional analysis before they can be fed into a machine to get an acceptable solution and thus require substantial human involvement to formulate a problem and ample skill and time to iteratively reformulate the problem until it is solvable by a machine. Thus, with the continuous human involvement, until the stability of the system, along with the state of the art tools

Earlier, Machine Learning models were used to successfully solve problems where final output was a simple function of input data, whereas in Deep Learning one can capture composite relations. As a result, it partly deals with learning data representations which are all about making things and presenting to real audiences.
In order to master the various aspects of Deep Learning one requires proficiency in core academic content, critical thinking, analytical thinking and self-directed learning. Deep Learning consists of mathematical models, which can be termed as a composition of the same type of functions wherein some of the functions can be changed so as to predict the final value.
Deep Learning is mathematical in nature and we have many real time applications showing its effectiveness like in image processing, restoration of colors in Black and white photos, pose estimation of a person and many other. Image processing problems generally include image restoration, enhancement, smoothing, estimation and filter design. In image processing, we partition a natural image into different pieces with geometric components or texture using various evolution approaches and functions. Further, image restoration is an active field of research which uses different mathematical model and techniques with real

But at the same time, these methods have shown their magnificent possibilities in mathematical research areas. In fact, under many circumstances, deep learning-based strategies have exceeded traditional mathematical approaches, which were once termed as state of the art for particular problems.
In recent times, many researchers from the mathematical community have decided to contribute in developing the in-depth mathematical foundation for deep learning by examining the problems from various angles and hence, have joined several theoretical computer scientists who are already on this pursuit.
Finally, the beauty of deep learning is its importance for mathematics and society as well as the richness of techniques which are involved, ranging from applied harmonic analysis and approximation theory over optimization methods to statistical learning