paper-with-me

홈 › Papers

Taming Overconfident Prediction on Unlabeled Data from Hindsight

2021-12-15 · Jing Li, Yuangang Pan, Ivor W. Tsang

Minimizing prediction uncertainty on unlabeled data is a key factor to achieve good performance in semi-supervised learning (SSL). The prediction uncertainty is typically expressed as the \emph{entropy} computed by the transformed probabilities in output space. Most existing works distill low-entropy prediction by either accepting the determining class (with the largest probability) as the true label or suppressing subtle predictions (with the smaller probabilities). Unarguably, these distillation strategies are usually heuristic and less informative for model training. From this discernment, this paper proposes a dual mechanism, named ADaptive Sharpening (\ADS), which first applies a soft-threshold to adaptively mask out determinate and negligible predictions, and then seamlessly sharpens the informed predictions, distilling certain predictions with the informed ones only. More importantly, we theoretically analyze the traits of \ADS by comparing with various distillation strategies. Numerous experiments verify that \ADS significantly improves the state-of-the-art SSL methods by making it a plug-in. Our proposed \ADS forges a cornerstone for future distillation-based SSL research.

📄 PDF Abstract BibTeX arXiv:2112.08200

Code (0)

등록된 구현이 없습니다.

Tasks

Prediction

Similar Papers 제목 키워드 기반

Do not trust what you trust: Miscalibration in Semi-supervised Learning

2024-03-22 · Shambhavi Mishra, Balamurali Murugesan, Ismail Ben Ayed, Marco Pedersoli 외

State-of-the-art semi-supervised learning (SSL) approaches rely on highly confident predictions to serve as pseudo-labels that guide the training on unlabeled samples. An inherent drawback of this strategy stems from the…

image-classificationImage Classification

Reducing Overconfident Errors outside the Known Distribution

2019-05-01 · ICLR 2019 5 · Zhizhong Li, Derek Hoiem

Intuitively, unfamiliarity should lead to lack of confidence. In reality, current algorithms often make highly confident yet wrong predictions when faced with unexpected test samples from an unknown distribution differen…

Domain AdaptationNovelty Detection

Path-level Hindsight Instructions for Semantic Exploration in Vision-Language Navigation

2026-07-02 · Sung June Kim, Sangpil Kim, Honglak Lee arxiv

On-policy exploration is a crucial component for training robust Vision-Language Navigation agents, as it exposes the policy to a broader state distribution. However, such exploration inevitably leads to trajectories tha…

Vision-Language Navigation

Reliable Graph Neural Networks for Drug Discovery Under Distributional Shift

2021-11-25 · Kehang Han, Balaji Lakshminarayanan, Jeremiah Liu

The concern of overconfident mis-predictions under distributional shift demands extensive reliability research on Graph Neural Networks used in critical tasks in drug discovery. Here we first introduce CardioTox, a real-…

Drug Discovery

CalibrateMix: Guided-Mixup Calibration of Image Semi-Supervised Models

2025-11-17 · Mehrab Mustafy Rahman, Jayanth Mohan, Tiberiu Sosea, Cornelia Caragea arxiv

Semi-supervised learning (SSL) has demonstrated high performance in image classification tasks by effectively utilizing both labeled and unlabeled data. However, existing SSL methods often suffer from poor calibration, w…

Image Classification