Active Learning for Convolutional Neural Networks: A Core-Set Approach
read the original abstract
Convolutional neural networks (CNNs) have been successfully applied to many recognition and learning tasks using a universal recipe; training a deep model on a very large dataset of supervised examples. However, this approach is rather restrictive in practice since collecting a large set of labeled images is very expensive. One way to ease this problem is coming up with smart ways for choosing images to be labelled from a very large collection (ie. active learning). Our empirical study suggests that many of the active learning heuristics in the literature are not effective when applied to CNNs in batch setting. Inspired by these limitations, we define the problem of active learning as core-set selection, ie. choosing set of points such that a model learned over the selected subset is competitive for the remaining data points. We further present a theoretical result characterizing the performance of any selected subset using the geometry of the datapoints. As an active learning algorithm, we choose the subset which is expected to yield best result according to our characterization. Our experiments show that the proposed method significantly outperforms existing approaches in image classification experiments by a large margin.
This paper has not been read by Pith yet.
Forward citations
Cited by 53 Pith papers
-
iSAGE: A Human-in-the-Loop Framework for Remote Sensing Semantic Segmentation via Sparse Point Supervision
iSAGE achieves near-dense mIoU performance in remote sensing semantic segmentation using iterative expert clicks on confident model errors with an error-weighted loss, using only 0.011-0.04% of pixels.
-
Bounded Behavioral Indistinguishability for Black-Box LLM Distillation
Introduces (ε,q,t,A)-behavioral indistinguishability and shows via Qwen/Llama experiments that LoRA distillation boosts semantic similarity but leaves detectable behavioral differences under adversarial evaluation.
-
Beyond What to Select: A Plug-and-play Oscillatory Data-Volume Scheduling for Efficient Model Training
PODS is a plug-and-play oscillatory data-volume scheduler that alternates low-ratio regularization phases with high-ratio recovery phases to improve data selection efficiency across training tasks.
-
EMA: Efficient Model Adaptation for Learning-based Systems
EMA cuts adaptation costs in learning-based systems by 14.9-42.4% and raises performance by 6.9-31.3% via state transformers for input alignment and prioritized high-utility data labeling.
-
MASS-DPO: Multi-negative Active Sample Selection for Direct Policy Optimization
MASS-DPO derives a Plackett-Luce-specific log-determinant Fisher information objective to select non-redundant negative samples, matching or exceeding multi-negative DPO performance with substantially fewer negatives ...
-
Active Testing of Large Language Models via Approximate Neyman Allocation
Proposes surrogate semantic entropy stratification followed by approximate Neyman allocation for active testing of LLMs on generative benchmarks, reporting up to 28% MSE reduction and 22.9% average budget savings vers...
-
Clip-level Uncertainty and Temporal-aware Active Learning for End-to-End Multi-Object Tracking
CUTAL scores multi-frame clips for uncertainty and enforces temporal diversity to train transformer MOT models to near full-supervision performance with 50% of the labels.
-
ContextualJailbreak: Evolutionary Red-Teaming via Simulated Conversational Priming
ContextualJailbreak uses evolutionary search over simulated primed dialogues with novel mutations to reach 90-100% attack success on open LLMs and transfers to some closed frontier models at 15-90% rates.
-
Dynamic Class-Aware Active Learning for Unbiased Satellite Image Segmentation
DCAU-AL is a new active learning strategy that dynamically weights samples by real-time class-wise segmentation performance gaps to improve per-class accuracy under imbalance in satellite imagery.
-
Positive-First Most Ambiguous: A Simple Active Learning Criterion for Interactive Retrieval of Rare Categories
PF-MA is a new active learning rule that favors likely-positive uncertain samples to speed up discovery of rare categories in imbalanced visual retrieval.
-
TableNet A Large-Scale Table Dataset with LLM-Powered Autonomous
TableNet is a new large-scale table dataset created via LLM multi-agent generation, combined with diversity-based active learning that achieves competitive performance on its test set and superior results on real-worl...
-
Towards Multimodal Active Learning: Efficient Learning with Limited Paired Data
Introduces the first active learning framework for unaligned multimodal data that selects alignments using uncertainty and diversity to cut annotation costs by up to 40% on benchmarks while preserving accuracy.
-
OD3: Optimization-free Dataset Distillation for Object Detection
OD3 presents an optimization-free dataset distillation framework for object detection that reports new state-of-the-art accuracy on COCO and VOC at compression ratios from 0.25% to 5%.
-
CRAFT: Cost-aware Refinement And Front-aware Tuning of Prompts
CRAFT is a Pareto-front prompt optimizer that allocates scarce LLM validation calls to candidates near the current front using accuracy- and cost-oriented generators plus NSGA-II retention.
-
Finding Needles in the Haystack: Transductive Active Labeling in Ecology
Transductive evaluation and a hybrid stopping criterion based on rarefaction curves improve rare-class discovery in long-tailed ecological active learning compared to standard inductive methods.
-
PHASER: Phase-Aware and Semantic Experience Replay for Vision-Language-Action Models
PHASER improves average success rate by up to 31% over uniform experience replay on LIBERO continual learning benchmarks for VLA models by phase-centric capacity allocation and semantic interference routing.
-
Single-Rollout Hidden-State Dynamics for Training-Free RLVR Data Selection
SHIFT selects compact RLVR training subsets using the magnitude of hidden-state change from a single inference rollout plus quality-weighted farthest-first coverage, outperforming training-free baselines on math reaso...
-
D3S2: Diffusion-Guided Dataset Distillation for Semantic Segmentation
D3S2 combines class-balanced mask selection with diffusion-guided image synthesis and two consistency losses to distill 1% datasets that yield 24.99% mIoU on ADE20K and 35.49% on COCO-Stuff, beating random selection.
-
Trajectory-Based Difficulty Scoring for Reliable Learning on Tabular Data
TDS uses per-tree prediction trajectories to derive instance difficulty scores that rank errors better than prior hardness measures and improve active learning, selective prediction, and Mondrian conformal prediction ...
-
Multimodal Distribution Matching for Vision-Language Dataset Distillation
MDM distills vision-language datasets via joint embedding clustering, weight-space model interpolation, and geometry-aware distribution matching on the unit hypersphere.
-
GoTTA be Diverse: Rethinking Memory Policies for Test-Time Adaptation
Diversity-aware memory policies improve test-time adaptation performance most under constrained memory budgets and challenging non-i.i.d. streams.
-
LiBaGS: Lightweight Boundary Gap Synthesis for Targeted Synthetic Data Selection
LiBaGS scores and selects synthetic data near decision boundaries using proximity, uncertainty, density, and validity, with boundary-gap allocation and marginal stopping to improve training accuracy.
-
Active Testing of Large Language Models via Approximate Neyman Allocation
Active testing via surrogate semantic entropy stratification and approximate Neyman allocation reduces MSE by up to 28% versus uniform sampling and saves about 23% of the labeling budget on language and multimodal benchmarks.
-
Gradient-Discrepancy Acquisition for Pool-Based Active Learning
A new gradient-discrepancy acquisition function derived from a generalization bound enables more effective pool-based active learning by selecting informative samples.
-
Gradient-Discrepancy Acquisition for Pool-Based Active Learning
Introduces gradient-discrepancy acquisition criterion derived from Luo et al. (2022) generalization bound for active learning.
-
Boundary-Centric Active Learning for Temporal Action Segmentation
B-ACT improves label efficiency in temporal action segmentation by selecting only boundary frames for annotation via a two-stage uncertainty-driven process that fuses neighborhood uncertainty, class ambiguity, and tem...
-
Select Smarter, Not More: Prompt-Aware Evaluation Scheduling with Submodular Guarantees
POES frames prompt evaluation as online adaptive testing and uses a provably submodular objective to pick informative examples, delivering 6.2% higher average accuracy and 35-60% token savings versus naive full-set scoring.
-
ADAM: A Systematic Data Extraction Attack on Agent Memory via Adaptive Querying
ADAM extracts data from LLM agent memory with up to 100% attack success rate by estimating data distribution and selecting queries via entropy guidance.
-
Scaling-Aware Data Selection for End-to-End Autonomous Driving Systems
MOSAIC is a scaling-aware data selection framework that outperforms baselines in training end-to-end autonomous driving planners, achieving comparable or better EPDMS scores with up to 80% less data.
-
Surprisingly High Redundancy in Electronic Structure Data Across Materials Explained by Low Intrinsic Dimensionality
Electronic structure datasets across materials show high redundancy from low intrinsic dimensionality, allowing pruning to 1/100th size with preserved chemical accuracy.
-
Rethinking Dataset Distillation for Classification: Do Distilled Sets Outperform Coresets?
Large-scale standardized benchmarks show state-of-the-art dataset distillation methods do not outperform coreset selection on ImageNet-scale data and have substantially higher construction costs.
-
OrderDP: A Theoretically Guaranteed Lossless Dynamic Data Pruning Framework
OrderDP is a plug-and-play data pruning method that selects a random subset then top-q samples to guarantee unbiased surrogate-loss training with convergence analysis and over 40% training cost reduction on CIFAR and ...
-
Finding Needles in the Haystack: Transductive Active Labeling in Ecology
Active learning evaluation in ecology should be transductive rather than inductive, with a hybrid stopping rule that combines prediction and discovery metrics to better recover long-tail classes.
-
An Efficient and Scalable Graph Condensation with Structure-Preserving
SP-ESGC decouples graph condensation into heat-kernel node condensation and pre-trained edge prediction for structure, claiming high efficiency and cross-GNN generalization on real-world datasets.
-
Are Candidate Models Really Needed for Active Learning?
Active learning with randomly initialized models achieves comparable results to traditional candidate-model methods, with low-confidence sampling proving most effective.
-
SLAP: Stratified Loss-based Pruning for On-Policy Data-Efficient Instruction Tuning
SLAP is a new batch-aware pruning framework that uses distribution-aware stratified sampling and Hessian-approximated gradients to select data, claiming 20-40% less data while matching or exceeding full-dataset perfor...
-
LiBaGS: Lightweight Boundary Gap Synthesis for Targeted Synthetic Data Selection
LiBaGS is a lightweight method that picks synthetic data near decision boundaries while checking density and validity to improve training accuracy over standard oversampling or uncertainty sampling.
-
Portable Active Learning for Object Detection
PAL is a portable active learning method for object detection that uses class-specific logistic classifiers for uncertainty and image-level diversity to select annotation batches, showing better label efficiency than ...
-
Evidence-based Decision Modeling for Synthetic Face Detection with Uncertainty-driven Active Learning
EMSFD uses Dirichlet-based evidence modeling to capture prediction uncertainty in synthetic face detection and applies uncertainty-driven active learning to achieve 15% higher accuracy than prior methods.
-
Evidence-based Decision Modeling for Synthetic Face Detection with Uncertainty-driven Active Learning
EMSFD models synthetic face detection via Dirichlet evidence and uncertainty-driven active learning, reporting 15% higher accuracy than prior state-of-the-art methods while improving reliability on out-of-distribution images.
-
Exploring and Exploiting Stability in Latent Flow Matching
Latent Flow Matching models exhibit inherent stability to data reduction and model shrinkage due to the flow matching objective, enabling reduced-dataset training and two-stage inference with over 2x speedup while pre...
-
Exploring and Exploiting Stability in Latent Flow Matching
LFM models exhibit stability to data reduction and capacity shrinkage that is tied to the flow matching objective, enabling reduced-data training and coarse-to-fine inference with over 2x speedup.
-
Uncertainty-Guided Edge Learning for Deep Image Regression in Remote Sensing
UGEL employs deep beta regression to estimate uncertainty in one forward pass, enabling faster convergence in edge learning for remote sensing image regression than active or semi-supervised baselines.
-
Selective Prediction from Agreement: A Lipschitz-Consistent Version Space Approach
Selective prediction abstains unless all Lipschitz-consistent heads in the version space agree on a certified label for each pool point.
-
When Active Learning Falls Short: An Empirical Study on Chemical Reaction Extraction
Active learning for chemical reaction extraction frequently produces non-monotonic learning curves and fails to deliver stable gains over random sampling because of strong pretraining, structured CRF decoding, and lab...
-
Neural Operator Representation of Granular Micromechanics-based Failure Envelope
A differentiable neural operator learns the mapping from granular microstructure configurations to failure envelopes, with physics-informed convexity enforcement and active learning for efficient training.
-
Labeled TrustSet Guided: Batch Active Learning with Reinforcement Learning
BRAL-T uses TrustSet-guided reinforcement learning for batch active learning and reports state-of-the-art results on 10 image classification benchmarks plus 2 fine-tuning tasks.
-
ALDEN: Boosting Private Data Extraction from Retrieval-Augmented Generation Systems via Active Learning and Distribution Estimation
ALDEN boosts private data extraction rates from RAG systems by combining active learning for query diversification with dynamic estimation of the underlying knowledge-base topic distribution.
-
Smart Picks in the Dark: Towards Efficient RLVR for Reasoning via Tracing Metacognitive Pivots
PivotTrace selects unlabeled data for RLVR by quantifying uncertainty via pivot density from attention dynamics, outperforming full supervision using only 29.3% annotations and converging 2.75 times faster.
-
Active Learning with Foundation Model Priors: Efficient Learning under Class Imbalance
Active learning with foundation model priors achieves over 50% annotation savings on imbalanced noisy datasets across image and text domains while maintaining performance.
-
Step-Video-T2V Technical Report: The Practice, Challenges, and Future of Video Foundation Model
Step-Video-T2V describes a 30B-parameter text-to-video model with custom Video-VAE, 3D DiT, flow matching, and Video-DPO that claims state-of-the-art results on a new internal benchmark.
-
ShieldGemma: Generative AI Content Moderation Based on Gemma
ShieldGemma delivers a family of Gemma2-based classifiers that outperform Llama Guard and WildCard on public safety benchmarks while introducing a synthetic-data curation pipeline for safety tasks.
-
Transformer-Based Active Learning for Data-Efficient Vaccine Epitope Selection in PRRS
Transformer models under active learning classify high-binding epitopes from a small docking dataset more accurately than random sampling or other architectures in low-data regimes for PRRS.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.