As this is scaled up, it has the potential to significantly broaden the range of problems to which quantum computers can be applied. Imagine a traditional computer that could only run computations of 20 computation steps, versus one that could run arbitrary length programs.
RESEARCH
-
Vision-Language Models Achieve SOTA Video Classification Performance
By
–
We use pre-trained vision-language models to improve video classification, achieving SOTA performance on Kinetics-400 and surpassing previous methods by 20-50% on five popular video datasets. #AAAI2023 Github: https://
github.com/whwu95/Text4Vis -
Scaling quantum error correction for longer computations
By
–
With further scaling, the error rates on "logical" qubits can hopefully be driven down sufficiently that we can run arbitrary-length quantum computations on a quantum computer, instead of ones with a limited number of steps.
-
Physical Qubits Combined Into Noise-Resistant Logical Qubits
By
–
What this work shows is that it is possible to group together many physical qubits and use all of their states to make a "logical" qubit, whose state is significantly less noisy (see the discussion in the blog post on surface codes, phase-flip errors and bit-flip errors).
-
Google Demonstrates Logical Qubit Prototype for Fault-Tolerant Computing
By
–
Developing useful large-scale quantum computers will require qubits with much lower error rates. Our latest results, published in @Nature
, demonstrate a prototype of a logical qubit, a milestone on the path toward scalable fault-tolerant quantum computing: https://
goo.gle/3KvE25V -
Physical Qubits Noise Limits Quantum Computation Steps
By
–
The reason this is important is that single physical qubits are noisy, and on a computation running on physical qubits, this probability of noise accumulates over time, and therefore you are limited to running programs that are no more than a modest handful of computation steps.
-
Google Quantum AI Achieves Logical Qubit Error Correction Breakthrough
By
–
Our @GoogleQuantumAI team has made an important step towards development of a large-scale quantum computer: building a prototype logical qubit & showing that quantum error correction makes that logical qubit better as it gets bigger. Congrats to the team! https://
ai.googleblog.com/2023/02/suppre
ssing-quantum-errors-by-scaling.html
… -
Underestimating Technology: From Internet to AI Revolution
By
–
En 1995, je découvre Internet. Mes profs en école de commerce : «ça ne marchera jamais». En 1998, je lance mon entreprise dans le web. Mon banquier : «ça ne marchera jamais». Et en 2023, un ministre du numérique compare une IA révolutionnaire à un perroquet. On est mal barrés.
-
Open Source Solution Replicates ChatGPT Training Process Efficiently
By
–
Open Source Solution Replicates ChatGPT Training Process! Ready To Go With Only 1.6GB GPU Memory And Gives You 7.73 Times Faster Training! https://
syncedreview.com/2023/02/22/ope
n-source-solution-replicates-chatgpt-training-process-ready-to-go-with-only-1-6gb-gpu-memory-and-gives-you-7-73-times-faster-training/
… -
ESG Technology Design and Sustainable Development Goals Implementation
By
–
Really appreciated indeed @Onalytica so close to heart and congrats to everyone here and supporting everything #ESG beyond as well, from embedding by #design & role of #tech through to consistent comparable measurement & beyond! #SDGs #TechForGood Many thanks Sally