Deep Learning is used heavily in autonomous cars for perception, image classification etc. What are they typical Deep Learning models that are used by Autonomous Cars and for which functionality?
Any related literature would also be very much appreciated.
Best,
Rasheed
Dear Prof. Hussein,
I think what cars and car networks lack today are genetic algorithms to solve the transportation algorithm for example using user data and that is why they are either underutilised or in disequilibrium systems service. Deep learning models should use our discovered "holomorphic resonance Dbranes String genetic protocols" (Mallick, Hamburger, Mallick (2016, 2017, 2018)), Mallick (2011)), which we have discovered but this research is at a fundamental Network Econophysics stage utilising String Theory computations.
Soumitra K. Mallick (CA)
for Soumitra K. Mallick, Nick Hamburger, Sandipan Mallick and others
RHMHM School of Mathematical Sciences Thinking
Dr. Hussain,
Research within image processing is very mature even as research continues. Deallong with advanced data within pixel, voxel etc. is still processing intensive. Hence, one algorithm type, such as genetic or other will not suffice to solve the entirety of what you ask. Additionally, performance within constrained knowledge economy environments has a long way to go To solve problems of real-time situational context and real inference which is still not yet possible with current methods. Much search is ongoing for diveloping higher fidelity relationships and causal models, as well as, machine statistical remembering which what ML really does but at highlet fidelity whether supervised or not. Bottom line is not one learning model is better than another. My suggestion from 30 years of doing this is to instead consider the highest fidelity needs for each Data type WRT your functional requirements. Work tirelessly on multiple data attack vectors and contextual understanding with the data...not just the Data...to truly Understand we must know in great detail and great accuracy the when, where, why, who, what, to get to the how. Lastly, I remember books out there describing what ML is best for what problem. Just realize it’s just one tiny part of a comprehensive solution. No magic bullet. Good luck!
Shanthi Rajaji is right. Usually, the researchers employ convoutional neural network model for autonomous car technology. There are many architectures of CNN that can be used.
Dear Rasheed
Try to use Convolution Neural Network (CNN), AlexNet, or GoogleNet
- Badges
- Science topic
- Similar topics
- Computer Science
- Images