Analog In-Memory Computing with Multilevel Memristive Devices for High Performance Computing
Open Compute Project via YouTube
The Most Addictive Python and SQL Courses
Learn the Skills Netflix, Meta, and Capital One Actually Hire For
Overview
Google, IBM & Meta Certificates – 40% Off
One plan covers every Professional Certificate on Coursera.
Unlock All Certificates
Watch a technical presentation exploring TetraMem's innovative solution to the von Neumann bottleneck through analog in-memory computing with multilevel memristive devices. Learn how pre-trained synaptic weights from cloud-based training are programmed directly into multi-level memristors/RRAMs for edge deployment, with capabilities for scenario-specific post-tuning. Discover the significance of achieving stable 8-bit and higher multi-level conductance in individual memristor devices, including breakthrough 11-bit/cell performance. Explore how these advances enable monolithically integrated semiconductor chips with large crossbar arrays on CMOS circuits, manufactured in commercial foundries for high-performance computing applications. Understand the latest developments in arbitrary precision computing based on analog computing, recently accepted for publication in a major journal.
Syllabus
Analog In-memory computing with multilevel memristive devices for high performance computing
Taught by
Open Compute Project