Personal tools

Difference between revisions of "Resilient Brain-Inspired Hyperdimensional Computing Architectures"

From iis-projects

Jump to: navigation, search
 
(3 intermediate revisions by one other user not shown)
Line 1: Line 1:
 
[[File:Brain_Architecture.jpg|thumb]]
 
[[File:Brain_Architecture.jpg|thumb]]
 
==Short Description==
 
==Short Description==
The mathematical properties of high-dimensional (HD) spaces show remarkable agreement with behaviors controlled by the brain. Computing with HD vectors, referred to as “hypervectors,” is a brain-inspired alternative to computing with numbers. HD computing is characterized by generality, scalability, and robustness against noise and variations in the computing platforms. Its tolerance for low-precision and faulty components is achieved by brain-inspired properties of hypervectors: (pseudo)randomness, high-dimensionality, and fully distributed holographic representations.
+
The mathematical properties of high-dimensional (HD) spaces show remarkable agreement with behaviors controlled by the brain. Computing with HD vectors, referred to as “hypervectors,” is a brain-inspired alternative to computing with numbers. The most important aspect of HD computing is its robustness against noise and variations in the computing platforms, allowing to implement resilient controllers. Its tolerance in low signal-to-noise ratio (SNR) conditions and for faulty components is achieved by brain-inspired properties of hypervectors: (pseudo)randomness, high-dimensionality, and fully distributed holographic representations.
  
In this project, your goal would be to develop an RTL implementation of an HD computing-based architecture and measure its resiliency against faulty components.   
+
In this project, your goal would be to design and develop an end-to-end robust HD processor with extremely resilient controller based on principles of HD computing, and measure its resiliency against faulty components.   
  
  
Line 33: Line 33:
  
 
[[#top|↑ top]]
 
[[#top|↑ top]]
 +
 
==Detailed Task Description==
 
==Detailed Task Description==
  
Line 53: Line 54:
 
[[#top|↑ top]]
 
[[#top|↑ top]]
  
[[Category:Neural_Processing]]
+
[[Category:Hyperdimensional Computing]]
 
[[Category:Digital]]
 
[[Category:Digital]]
 
     [[Category:ASIC]]
 
     [[Category:ASIC]]
Line 59: Line 60:
 
[[Category:Available]]
 
[[Category:Available]]
 
[[Category:Semester Thesis]]
 
[[Category:Semester Thesis]]
 +
[[Category:2018]]
 
[[Category:Master Thesis]]
 
[[Category:Master Thesis]]
  

Latest revision as of 18:08, 29 January 2021

Brain Architecture.jpg

Short Description

The mathematical properties of high-dimensional (HD) spaces show remarkable agreement with behaviors controlled by the brain. Computing with HD vectors, referred to as “hypervectors,” is a brain-inspired alternative to computing with numbers. The most important aspect of HD computing is its robustness against noise and variations in the computing platforms, allowing to implement resilient controllers. Its tolerance in low signal-to-noise ratio (SNR) conditions and for faulty components is achieved by brain-inspired properties of hypervectors: (pseudo)randomness, high-dimensionality, and fully distributed holographic representations.

In this project, your goal would be to design and develop an end-to-end robust HD processor with extremely resilient controller based on principles of HD computing, and measure its resiliency against faulty components.


Status: Available

Looking for 1-2 Semester/Master students
Contact: Abbas Rahimi

Prerequisites

HDL coding
VLSI I
Fault Injection and Testing

Character

20% Theory
40% Architecture Design
40% Test

Professor

Luca Benini

↑ top

Detailed Task Description

Goals

Practical Details

Results

Links

↑ top