Dual Certified White-Box Inference for Input Convex Neural Networks
Quick summary
arXiv:2605.04722v2 Announce Type: replace-cross Abstract: Input convex neural networks (ICNNs) are used to learn convex objectives whose minimizers define decisions, making efficient and reliable optimization central to inference. At nonsmooth inputs, automatic differentiation returns a single derivative rather than the full subdifferential governing optimality and descent. Second-order cone ICNNs (SOC-ICNNs) admit an exact representation as value functions of parametric second-order cone programs, providing a white-box approach to recovering their full subdifferentials from optimal dual multi
Key takeaways
- arXiv:2605.04722v2 Announce Type: replace-cross Abstract: Input convex neural networks (ICNNs) are used to learn convex objectives whose minimizers define decisions, making efficient and reliable optimization central to inference.
- At nonsmooth inputs, automatic differentiation returns a single derivative rather than the full subdifferential governing optimality and descent.
- Second-order cone ICNNs (SOC-ICNNs) admit an exact representation as value functions of parametric second-order cone programs, providing a white-box approach to recovering their full subdifferentials from optimal dual multi
Why it matters
The importance of “Dual Certified White-Box Inference for Input Convex Neural Networks” will be measured by what changes in practice. User behavior, access conditions, verifiable performance and responsible-use outcomes are the signals worth following.
Member comments