presentation jaswin.ppt Ascr is a four layer
- 1. Prototyping Presentation The best marketing tool for your business. PRESENTED BY : JASWINRAJ.K
- 2. INTRODUCTION Neuromorphic engineering is a multidisciplinary field that designs artificial neural systems to mimic the human brain’s functions. It’s also known as neuromorphic computing.
- 3. HISTORY Early Concepts : (1940s-1970s) First computational model of a neuron was introduced Birth of Neuromorphic Engineering : (1980s) Carver Mead coined the term "neuromorphic engineering" Early Neuromorphic Chips : (1990s) Development of electronic Cochlea using CMOS VLSI and Silicon Retina Advancements in Hardware and Algorithm Research : (2000) Blue Brain Project Started, Georgia Tech presented a Field Programmable Neural Array Breakthroughs and Commercialization : (2010) Development of Neuromorphic Chips: IBM TrueNorth, Intel Loihi, DynapCNN, Odin, DYNAP-SE2, etc. Expansion and Growth : (2020) Loihi 2, SpiNNaker 2, Akida, BrainScaleS 2, ReckOn, Speck, Xylo, Spiking Neural Processor T1:
- 4. HOW NEUROMORPHIC ENGINEERING WORKS Neuromorphic engineering incorporates several key characteristics, including massive parallelism, event-driven computation, and the use of spiking neural networks. These features enable neuromorphic systems to process information in a manner that is fundamentally different from traditional von Neumann computers. By mirroring the brain’s neural processing, neuromorphic engineering strives to create systems that can perceive, act, and learn from their environment autonomously.
- 5. USES OF NEUROMORPHIC ENGINEERING Neuromorphic engineering is an interdisciplinary field that uses biology, physics, mathematics, computer science, and electronic engineering to design artificial neural systems. These systems are inspired by the human brain and are used in a variety of applications, including
- 6. Computing :Neuromorphic engineering can be used to develop new computing technologies that are complementary to digital computing devices. These technologies include spiking neural networks, multineuron chips, and large-scale artificial neural systems. Machine learning: Neuromorphic computing can be used in machine learning applications to recognize patterns in speech and natural language, analyze medical images, and process imaging signals from brain scans. Robotics: Neuromorphic computing can be used to enhance a robot's real-time learning and decision- making skills, Research and defense: Neuromorphic processors are mostly intended for research and defense purposes.
- 8. ADVANTAGES Energy Efficiency: Minimizes power usage like the brain, leading to significant energy savings. Real-time Processing: Processes data and makes decisions quickly, useful for autonomous systems. Adaptability: Learns and improves from new data without extensive reprogramming. Robustness: Offers greater fault tolerance and reliability by mimicking neural networks. Scalability: Efficiently handles large data and complex tasks with its brain- inspired design.
- 9. DISADVANTAGE Design Complexity: Creating neuromorphic systems is intricate and costly, requiring advanced knowledge and leading to longer development times. Scalability Issues: Replicating the brain’s vast neural network in a practical and efficient manner is difficult. Power Consumption: Achieving energy efficiency comparable to the human brain is challenging, and power use can still be high. Software Limitations: Limited software and tools make it hard to fully leverage neuromorphic hardware. Integration Problems: Combining neuromorphic systems with existing technologies can be complex and problematic. Maturity: The field is still evolving, with technology and applications not as developed as traditional computing.
- 10. ANY DOUBTS ?
- 11. THANK YOU