PORTALE DELLA DIDATTICA

PORTALE DELLA DIDATTICA

PORTALE DELLA DIDATTICA

Elenco notifiche



Computing Paradigms for Error-Tolerant Applications

01UJRIU

A.A. 2020/21

Course Language

Inglese

Degree programme(s)

Doctorate Research in Ingegneria Informatica E Dei Sistemi - Torino

Course structure
Teaching Hours
Lezioni 25
Lecturers
Teacher Status SSD h.Les h.Ex h.Lab h.Tut Years teaching
Jahier Pagliari Daniele   Ricercatore a tempo det. L.240/10 art.24-B IINF-05/A 15 0 0 0 2
Co-lectures
Espandi

Context
SSD CFU Activities Area context
*** N/A ***    
PERIOD: MARCH - JULY The days of computer speeds doubling every 18 months, not caring about energy consumption, are long over. For this reason, researchers are looking for alternatives to making computing platforms fast and energy efficient. This course will introduce Approximate Computing and Neuromorphic Computing, two of the most relevant new computation paradigms that have arisen in recent years to cope with the increasing need for energy efficiency and high performance of modern applications. Both these paradigms leverage the fact that many applications (machine learning, multimedia. robotics, etc.) are tolerant to errors, and modify the structure of computing platforms accordingly, at both hardware and software levels, in order to increase performance or energy efficiency. The course will provide an introductory overview of the main Approximate and Neuromorphic Computing techniques proposed by researchers in academia and industry, focusing both on hardware architectures and on software stacks. The goal is to hopefully provide the audience with new tools that they can use in their own research whenever they will be dealing with error-tolerant and/or event-driven applications.
PERIOD: MARCH - JULY The days of computer speeds doubling every 18 months, not caring about energy consumption, are long over. For this reason, researchers are looking for alternatives to making computing platforms fast and energy efficient. This course will introduce Approximate Computing and Neuromorphic Computing, two of the most relevant new computation paradigms that have arisen in recent years to cope with the increasing need for energy efficiency and high performance of modern applications. Both these paradigms leverage the fact that many applications (machine learning, multimedia. robotics, etc.) are tolerant to errors, and modify the structure of computing platforms accordingly, at both hardware and software levels, in order to increase performance or energy efficiency. The course will provide an introductory overview of the main Approximate and Neuromorphic Computing techniques proposed by researchers in academia and industry, focusing both on hardware architectures and on software stacks. The goal is to hopefully provide the audience with new tools that they can use in their own research whenever they will be dealing with error-tolerant and/or event-driven applications.
A minimum knowledge in the following fields is required to enjoy the course: - Computer architecture - Hardware design/Electronic Design Automation - Embedded systems - Embedded software/firmware - Parallel software - Low power design and energy optimization for digital systems
A minimum knowledge in the following fields is required to enjoy the course: - Computer architecture - Hardware design/Electronic Design Automation - Embedded systems - Embedded software/firmware - Parallel software - Low power design and energy optimization for digital systems
- Introduction: new paradigms for post-Moore’s Law computing - Error tolerance in applications - Approximate Computing: general concepts - Approximate Computing: examples of applications - Computing with Spiking Neurons: general concepts - Computing with Spiking Neurons: example of applications - Future trends and challenges The final examination consists in a presentation of the student showing how one or more of the techniques introduced in the course can be applied to his/her own research.
- Introduction: new paradigms for post-Moore’s Law computing - Error tolerance in applications - Approximate Computing: general concepts - Approximate Computing: examples of applications - Computing with Spiking Neurons: general concepts - Computing with Spiking Neurons: example of applications - Future trends and challenges The final examination consists in a presentation of the student showing how one or more of the techniques introduced in the course can be applied to his/her own research.
In presenza
On site
Presentazione orale
Oral presentation
P.D.2-2 - Marzo
P.D.2-2 - March