Energy Efficient AI – Page 3 – MIT AI Hardware Program

Rendering shows several layers, including a metallic block on bottom. Above this block are lattices of layered atoms. Above these lattices, a twist of energy has a two-sided arrow, with the top part emphasized.

Researchers Harness 2D Magnetic Materials for Energy-efficient Computing

Adam Zewe | MIT News

An MIT team precisely controlled an ultrathin magnet at room temperature, which could enable faster, more efficient processors and computer memories.

Energy Efficient AIHigh Performance Computation

Artist’s concept of four domes comprised of spheres representing nanoparticles. Three of the domes have different colored particle beams shining down on them.

Team Engineers Nanoparticles Using Ion Irradiation to Advance Clean Energy and Fuel Conversion

Elizabeth Thomson | Materials Research Laboratory

Combining the techniques, metal exsolution and ion irradiation, demonstrates control over key nanoparticle properties leading to better performance.

Energy Efficient AIAI HardwareMaterials

Analog In-Memory Computing for Deep Learning Inference

Wednesday, November 15, 2023 | 12:00 - 1:00pm ET
Hybrid

Grier A (34-401A)
50 Vassar Street Cambridge, MA

Energy Efficient AINeuromorphic Computing

Rendering shows 3 parts as if on a grey table: a white model house on top; a fuel cell sandwiched in between two metal plates with spherical molecules floating around it; and on bottom is the electrolyzer, which looks similar to the fuel cell and has molecules floating around it.

Engineers Develop an Efficient Process to Make Fuel from Carbon Dioxide

David L. Chandler | MIT News

The approach directly converts the greenhouse gas into formate, a solid fuel that can be stored indefinitely and could be used to heat homes or power industries.

Energy Efficient AIAI HardwareMaterials

In the foreground is a search bar with a magnifying glass icon and a lock icon. In a background are blueprints, a neural network made up of spheres and bars, and more copies of the search bar.

Accelerating AI Tasks While Preserving Data Security

Adam Zewe | MIT News

The SecureLoop search tool efficiently identifies secure designs for hardware that can boost the performance of complex AI tasks, while requiring less energy.

Energy Efficient AI

Colorful abstract drawing with blue lightning streaks on the left, gold microcircuits on the right, and a computer chip with a lighting bolt in the middle

System Combines Light and Electrons to Unlock Faster, Greener Computing

Alex Shipps | MIT CSAIL

“Lightning” system connects photons to the electronic components of computers using a novel abstraction, creating the first photonic computing prototype to serve real-time machine-learning inference requests.

Energy Efficient AIHigh Performance ComputationPhotonics

Realistic drawing of a printed circuit board, with circuit lines projecting above it in the shape of a brain

Machine-learning System Based on Light Could Yield More Powerful, Efficient Large Language Models

Elizabeth A. Thomson | Materials Research Laboratory

MIT system demonstrates greater than 100-fold improvement in energy efficiency and a 25-fold improvement in compute density compared with current systems.

Neural NetworksEnergy Efficient AIPhotonics