DeepSeek pitches new route to scale AI, but researchers call for more testing
DeepSeek’s proposed ‘mHC’ design could change how AI models are trained, but experts caution it still needs to prove itself at scale
However, experts cautioned that while the approach could prove far-reaching, it might still prove difficult to put into practice.
HC was developed to address limitations of Residual Networks (ResNet), an architecture that underpins many modern deep-learning models, including LLMs.
ResNet was proposed about a decade ago by four researchers at Microsoft Research Asia, including prominent computer scientist Kaiming He.
DeepSeek’s paper marks the Chinese AI start-up’s latest effort to improve model training efficiency with limited computing resources, fuelling speculation that its next models could incorporate the new architecture.