Pro@programming.dev to Programming@programming.devEnglish · 6 days agoSurprisingly Fast AI-Generated Kernels We Didn’t Mean to Publish (Yet)crfm.stanford.eduexternal-linkmessage-square5fedilinkarrow-up122arrow-down16cross-posted to: hackernews
arrow-up116arrow-down1external-linkSurprisingly Fast AI-Generated Kernels We Didn’t Mean to Publish (Yet)crfm.stanford.eduPro@programming.dev to Programming@programming.devEnglish · 6 days agomessage-square5fedilinkcross-posted to: hackernews
minus-squareonlinepersona@programming.devlinkfedilinkarrow-up1·4 days agoThe weights for the neural network or the embeddings? Anti Commercial-AI license
minus-squarebitfucker@programming.devlinkfedilinkarrow-up3·4 days agoNo. Normally the kernel doesn’t get updated in the network during training. They are called hyper-parameters. They do affect training, but they are not updated by the training algorithm
The weights for the neural network or the embeddings?
Anti Commercial-AI license
No. Normally the kernel doesn’t get updated in the network during training. They are called hyper-parameters. They do affect training, but they are not updated by the training algorithm