Pith. sign in

REVIEW 1 cited by

Balanced Meta-Softmax for Long-Tailed Visual Recognition

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2007.10740 v3 pith:MHSNO744 submitted 2020-07-21 cs.LG cs.CVstat.ML

Balanced Meta-Softmax for Long-Tailed Visual Recognition

classification cs.LG cs.CVstat.ML
keywords long-tailedbalancedsoftmaxmeta-softmaxrecognitionvisualboundclassification
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

Deep classifiers have achieved great success in visual recognition. However, real-world data is long-tailed by nature, leading to the mismatch between training and testing distributions. In this paper, we show that the Softmax function, though used in most classification tasks, gives a biased gradient estimation under the long-tailed setup. This paper presents Balanced Softmax, an elegant unbiased extension of Softmax, to accommodate the label distribution shift between training and testing. Theoretically, we derive the generalization bound for multiclass Softmax regression and show our loss minimizes the bound. In addition, we introduce Balanced Meta-Softmax, applying a complementary Meta Sampler to estimate the optimal class sample rate and further improve long-tailed learning. In our experiments, we demonstrate that Balanced Meta-Softmax outperforms state-of-the-art long-tailed classification solutions on both visual recognition and instance segmentation tasks.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. On the Impact of Class Imbalance on the Learning Dynamics of Deep Neural Networks:An Intuitive Insight

    cs.LG 2026-05 unverdicted novelty 3.0

    Class imbalance causes DNNs to underfit minority classes early in training and produce non-generalizable minority representations later by overfitting to minimize overall loss.