Eigenvalues of the Hessian in Deep Learning: The Origin of Symmetry and Its Breaking
A symmetry-breaking account explains clustered Hessian spectra in trained deep networks, including ReLU, CNN, graph, and transformer models.
3 stories, newest first · one card per story
A symmetry-breaking account explains clustered Hessian spectra in trained deep networks, including ReLU, CNN, graph, and transformer models.
G2PTQ refreshes gradients and Hessians during LLM quantization, beating GPTQ-style baselines.
0day Rubbish disclosed an unauthenticated Hessian deserialization flaw in Accurate Online Private Cloud on-prem allowing JNDI remote class loading, rated 9.8.
Today 1.6K readers · 2.4K views · 775 reads · all time 189K views