Artificial Intelligence thread
- Thread starter 9dashline
- Start date
this is kind of interesting. Now that Huawei came out with Xiaoyi, everyone else is going to have to talk about it even if what they have is not on the same level or worth talking about yet.Lei Jun: The 1.3 billion parameter model has already run through the mobile phone
Jiwei.com reported that on the evening of August 14th, Lei Jun, chairman of Xiaomi, said in his annual speech that Xiaomi fully embraces large models. The latest large model with 1.3 billion parameters has successfully run locally on the mobile phone, and some scenes can be compared to the 6 billion parameter model. The result of running in the cloud. Xiao Ai has upgraded the large model, and invites to test it today.
Lei Jun said that Xiaomi’s R&D investment in 2023 is expected to reach 20 billion. Currently, the number of global patent authorizations is 32,000+, and the 5G standard essential patent statement is valid for 4.1% of the global patent families. Xiaomi has laid out 12 technical fields and 99 subdivided tracks, and will invest 100 billion yuan in technology in the next five years. In terms of AI, Xiaomi has been laying out since July 2016, and established a large-scale model team in April this year, with more than 3,000 AI-related teams.
The European Centre for Medium-Range Weather Forecasts is working with Chinese tech giant Huawei to launch an AI-based forecasting system
An intergovernmental organisation in Europe is working closely with a Chinese tech giant to apply (AI) to weather forecasting.
The European Centre for Medium-Range Weather Forecasts launched a Chinese AI model capable of predicting global weather on its website late last month.
The model, Pangu-Weather, was developed by .
The has seen the Chinese tech company translate the science behind the weather into practical applications, while the European weather agency has begun to embrace AI in its daily forecasts.
In an exclusive interview with the South China Morning Post on August 4, Huawei’s chief scientist for AI, Tian Qi, discussed the cooperation between the two organisations, as well as the potential for in meteorology.
“The decision of including Pangu-Weather in their forecasting is partly due to the model’s high performance and the centre recognising AI’s in this field,” Tian said.
An intergovernmental organisation in Europe is working closely with a Chinese tech giant to apply (AI) to weather forecasting.
The European Centre for Medium-Range Weather Forecasts launched a Chinese AI model capable of predicting global weather on its website late last month.
The model, Pangu-Weather, was developed by .
The has seen the Chinese tech company translate the science behind the weather into practical applications, while the European weather agency has begun to embrace AI in its daily forecasts.
In an exclusive interview with the South China Morning Post on August 4, Huawei’s chief scientist for AI, Tian Qi, discussed the cooperation between the two organisations, as well as the potential for in meteorology.
“The decision of including Pangu-Weather in their forecasting is partly due to the model’s high performance and the centre recognising AI’s in this field,” Tian said.
I would bet in 3 years IBM will report that they have a severe staff shortages and it was a big mistake.
After watching their presentation, I think the voice assistance is quite impressive in understanding+carrying out voice commands. This is expected to help schools & businesses.
The ability to look at a picture and create & tell a story is also impressive. Could be great for families reading books to kids
Spark Model can create AIGC contents in perfect English. Would be ideal for advertising from corporations & travel agencies
It also has a language teacher that can be used to learn English & Chinese. In general, iFlyTek showcased how AI tools can be used as education tools.
The AI machine they came out is also quite interesting. They really emphasized about it being fully domestic, using domestic technology (Huawei, ascend, Mindspore, high speed comunication & distributed storage combined with iFlyTek’s LLM training platform) and cannot be sanctioned. It’s kind of like iFlyTek’s A100 DGX.
this AI assistant thing is a huge deal imo.
We are going to start seeing it in people's homes soon enough
China used a homegrown -assisted system for the first time in heat detection of ultra-high voltage (UHV) transmission lines. With the help of infrared thermograph tech, heat fluctuations & fault points can be identified in time with 90%+ accuracy rate, reducing manual workload..
An open source AI and brain simulation platform from CAS.
Paper:
News story:
8月15日，记者从中国科学院自动化研究所获悉，该所研究人员成功打造全脉冲神经网络类脑认知智能引擎“智脉”，并将其全面开源开放。该平台将为探索面向通用人工智能的类脑智能研究提供基础支撑，助力探索自然智能的计算本质和新一代人工智能的发展。相关研究成果以封面文章的形式发表于细胞出版社旗下《Patterns》期刊。
据悉，“智脉”是一个基于类脑脉冲神经网络的人工智能与脑模拟计算平台。它以多尺度生物可塑性原理为基础，支持全脉冲神经网络建模，具备脑启发的人工智能模型以及脑功能和结构模拟能力。
论文通讯作者、中国科学院自动化所研究员曾毅认为，只有充分借鉴人脑结构及其智能机理以及自然演化的规律与机制，才能实现真正意义且对人类有益的人工智能。
目前，“智脉”为类脑人工智能和计算神经科学的研究者提供了一套相对完整的、系统化的接口组件，包括了不同精细程度的神经元计算模型，丰富的类脑学习与可塑性法则，不同的神经网络连接模式，多样的编码方式，丰富的功能性脑区模型以及软硬件协同系统。这为研发基于脉冲神经网络的人工智能和计算神经科学研究提供了强有力的基础设施。
Paper:
BrainCog: A spiking neural network based, brain-inspired cognitive intelligence engine for brain-inspired AI and brain simulation
Highlights
- BrainCog is an integrated platform for brain-inspired AI and brain simulation
- BrainCog offers a variety of universal foundational components
- BrainCog supports diverse cognitive functions in various domains
- BORN integrates BrainCog’s cognitive functions to create advanced AI
The bigger picture
To integrate the scientific exploration of the nature of biological intelligence and develop artificial intelligence (AI), the research community urgently requires an open-source platform that can simultaneously support brain-inspired AI and brain simulation across multiple scales. To meet this need, we release a spiking neural network (SNN)-based, brain-inspired cognitive intelligence engine (BrainCog) that provides a variety of universal essential components, including spiking neurons, encoding strategies, learning rules, brain areas, and hardware-software co-design. With these easy-to-use components, BrainCog incorporates numerous brain-inspired AI models that cover five categories of brain-inspired cognitive functions. It also supports multi-scale brain structure and function simulation. We also provide BORN, an SNN-driven, brain-inspired AI engine that integrates multiple BrainCog components and cognitive functions to build advanced AI models and robotics applications.News story:
8月15日，记者从中国科学院自动化研究所获悉，该所研究人员成功打造全脉冲神经网络类脑认知智能引擎“智脉”，并将其全面开源开放。该平台将为探索面向通用人工智能的类脑智能研究提供基础支撑，助力探索自然智能的计算本质和新一代人工智能的发展。相关研究成果以封面文章的形式发表于细胞出版社旗下《Patterns》期刊。
据悉，“智脉”是一个基于类脑脉冲神经网络的人工智能与脑模拟计算平台。它以多尺度生物可塑性原理为基础，支持全脉冲神经网络建模，具备脑启发的人工智能模型以及脑功能和结构模拟能力。
论文通讯作者、中国科学院自动化所研究员曾毅认为，只有充分借鉴人脑结构及其智能机理以及自然演化的规律与机制，才能实现真正意义且对人类有益的人工智能。
目前，“智脉”为类脑人工智能和计算神经科学的研究者提供了一套相对完整的、系统化的接口组件，包括了不同精细程度的神经元计算模型，丰富的类脑学习与可塑性法则，不同的神经网络连接模式，多样的编码方式，丰富的功能性脑区模型以及软硬件协同系统。这为研发基于脉冲神经网络的人工智能和计算神经科学研究提供了强有力的基础设施。