AI Paper Atlas

Research map / Learning theory and algorithms / Prediction and decision losses

In Defense of Softmax Parametrization for Calibrated and Consistent Learning to Defer

Yuzhou Cao, Hussein Mozannar, Lei Feng, Hongxin Wei, Bo An

NeurIPS 2023 · Learning theory and algorithms › Prediction and decision losses · approach: Losses for decision-making · 1 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use asymmetric softmax surrogate (A-SM) instead of symmetric softmax (S-SM)L2D surrogate · CIFAR100 expert acc 94%, k=20coverage (%): 77.72 -> 98.16Improved
Use A-SM instead of S-SM with a high-accuracy expertL2D surrogate · CIFAR100 expert acc 94%, k=60error (%): 18.69 -> 19.30No clear effect
Use A-SM instead of S-SM with a 75% accuracy expertL2D surrogate · CIFAR100 expert acc 75%, k=60error (%): 24.05 -> 22.48Improved
Use A-SM instead of S-SML2D surrogate · CIFAR100 expert acc 94%, k=60ECE (%): 11.10 -> 5.96Improved
Use A-SM instead of A-OvAL2D surrogate · CIFAR100 expert acc 94%, k=40coverage (%): 80.46 -> 92.20Improved
Use A-SM instead of A-OvAL2D surrogate · CIFAR100 expert acc 94%, k=40ECE (%): 6.81 -> 6.58No clear effect
Use bounded asymmetric OvA instead of symmetric OvA (S-OvA)L2D surrogate · CIFAR100 expert acc 94%, k=20coverage (%): 85.40 -> 90.40Improved
Use A-SM instead of S-SM under 10% deferral budgetL2D surrogate · CIFAR100 expert acc 94%, k=40budgeted error (%): 46.69 -> 22.17Improved
Use A-SM instead of A-OvA under 10% deferral budgetL2D surrogate · HateSpeechbudgeted error (%): 16.525 -> 15.07Improved
Use A-SM instead of S-SML2D surrogate · HateSpeecherror (%): 8.65 -> 8.06Improved
Use A-SM instead of A-OvAL2D surrogate · ImageNet-16H noise 095error (%): 14.17 -> 12.59Improved
Use A-SM instead of A-OvAL2D surrogate · CIFAR10HECE (%): 0.95 -> 0.94No clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy