Janhavi H Borse¹ and Dipdf/ASCSup>2*

¹Research Scholar, SKNCOE, SPPU, Pune, India
²Associate Professor, MKSSS's Cummins College of Engineering, Pune, India

*Corresponding Author: Dipti D Patil, Associate Professor, MKSSS's Cummins College of Engineering, Pune, India.

Received: October 22, 2021; Published: January 31, 2022

Earlier trends in computer science introduced symbolic AI as a giant leap in machine intelligence. Emphasis was on how machines can be programmed to perform human tasks. Symbolic AI is a program written which takes specific inputs and produces the intended outputs using a set of pre-determined rules. It makes symbolic AI a task-specific program, and we need to write a different set of rules for the different tasks. Gradually researchers shifted their focus from task-specific programs to define learning models that can be used for various tasks. A subfield of AI, called Machine Learning, started emerging where the focus is on training the models for knowing the patterns hidden in the data. These patterns are nothing but the rules that can be used for making predictions on unknown data samples. With the advent of the internet, a tremendous amount of data started generating and increased processing capabilities. It forced research communities to dive into deep neural networks and invent newer, more sophisticated neural models to understand the data more accurately. A deep neural network does so by providing detailed representations of the input data using its layered architecture. At each next layer, more details are analyzed and represented. Thus deep neural networks are a sub-domain of machine learning with much more capabilities.

Citation

Citation: Janhavi H Borse and Dipti D Patil. “Impact of Deep Neural Learning in Computer Science". Acta Scientific Computer Sciences 4.2 (2022): 56-57.

Copyright

Copyright: © 2022 Janhavi H Borse1 and Dipti D Patil. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.