binary classification
A type of classification task where two distinct classes are predicted. Binary classification models output a single probability score indicating membership in one of the two classes, commonly used in tasks like spam detection.
- Attention with Trained Embeddings Provably Selects Important Tokens
- Balancing Positive and Negative Classification Error Rates in Positive-Unlabeled Learning
- DETree: DEtecting Human-AI Collaborative Texts via Tree-Structured Hierarchical Representation Learning
- DIsoN: Decentralized Isolation Networks for Out-of-Distribution Detection in Medical Imaging
- Efficient PAC Learning for Realizable-Statistic Models via Convex Surrogates
- Estimating Model Performance Under Covariate Shift Without Labels
- How to Learn a Star: Binary Classification with Starshaped Polyhedral Sets
- Human Texts Are Outliers: Detecting LLM-generated Texts via Out-of-distribution Detection
- Neural Rule Lists: Learning Discretizations, Rules, and Order in One Go
- On Agnostic PAC Learning in the Small Error Regime
- QuanDA: Quantile-Based Discriminant Analysis for High-Dimensional Imbalanced Classification
- SPACE: Noise Contrastive Estimation Stabilizes Self-Play Fine-Tuning for Large Language Models
- Self-Boost via Optimal Retraining: An Analysis via Approximate Message Passing
- The Effect of Optimal Self-Distillation in Noisy Gaussian Mixture Model
- The Rich and the Simple: On the Implicit Bias of Adam and SGD
- Zeroth-Order Optimization Finds Flat Minima