# How Much More Data Do I Need? Estimating Requirements for Downstream Tasks

arXiv: https://arxiv.org/abs/2207.01725 · HTML: https://mlsanity.com/paper/2207.01725/

## Key facts

- Cited by 23 papers in the map (data through 14 Oct 2026).
- Builds on 28 of its 43 references that are in the map.
- First posted on arXiv 4 Jul 2022.

## Builds on (most cited first, 28 of 28)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [nuScenes: A multimodal dataset for autonomous driving](https://mlsanity.com/paper/1903.11027/) (2019)
- [SSD: Single Shot MultiBox Detector](https://mlsanity.com/paper/1512.02325/) (2015)
- [Wide Residual Networks](https://mlsanity.com/paper/1605.07146/) (2016)
- [Rethinking Atrous Convolution for Semantic Image Segmentation](https://mlsanity.com/paper/1706.05587/) (2017)
- [BDD100K: A Diverse Driving Dataset for Heterogeneous Multitask Learning](https://mlsanity.com/paper/1805.04687/) (2018)
- [Lift, Splat, Shoot: Encoding Images From Arbitrary Camera Rigs by Implicitly Unprojecting to 3D](https://mlsanity.com/paper/2008.05711/) (2020)
- [Scaling Vision Transformers](https://mlsanity.com/paper/2106.04560/) (2021)
- [Revisiting Unreasonable Effectiveness of Data in Deep Learning Era](https://mlsanity.com/paper/1707.02968/) (2017)
- [Deep Learning Scaling is Predictable, Empirically](https://mlsanity.com/paper/1712.00409/) (2017)
- [FCOS3D: Fully Convolutional One-Stage Monocular 3D Object Detection](https://mlsanity.com/paper/2104.10956/) (2021)
- [Explaining Neural Scaling Laws](https://mlsanity.com/paper/2102.06701/) (2021)
- [Selection via Proxy: Efficient Data Selection for Deep Learning](https://mlsanity.com/paper/1906.11829/) (2019)
- [Learning Loss for Active Learning](https://mlsanity.com/paper/1905.03677/) (2019)
- [Variational Adversarial Active Learning](https://mlsanity.com/paper/1904.00370/) (2019)
- [A Constructive Prediction of the Generalization Error Across Scales](https://mlsanity.com/paper/1909.12673/) (2019)
- [Efficient Interactive Annotation of Segmentation Datasets with Polygon-RNN++](https://mlsanity.com/paper/1803.09693/) (2018)
- [Exploring the Limits of Large Scale Pre-training](https://mlsanity.com/paper/2110.02095/) (2021)
- [Learning Curve Theory](https://mlsanity.com/paper/2102.04074/) (2021)
- [The Shape of Learning Curves: a Review](https://mlsanity.com/paper/2103.10948/) (2021)
- [NeurIPS 2020 Competition: Predicting Generalization in Deep Learning](https://mlsanity.com/paper/2012.07976/) (2020)
- [Low Budget Active Learning via Wasserstein Distance: An Integer Programming Approach](https://mlsanity.com/paper/2106.02968/) (2021)
- [Not All Labels Are Equal: Rationalizing The Labeling Costs for Training Object Detection](https://mlsanity.com/paper/2106.11921/) (2021)
- [Learning Curves for Analysis of Deep Networks](https://mlsanity.com/paper/2010.11029/) (2020)
- [Low-Shot Validation: Active Importance Sampling for Estimating Classifier Performance on Rare Categories](https://mlsanity.com/paper/2109.05720/) (2021)
- [A Theoretical-Empirical Approach to Estimating Sample Complexity of DNNs](https://mlsanity.com/paper/2105.01867/) (2021)

## Built on by (most cited first, 23 of 23)

- [Playground v2.5: Three Insights towards Enhancing Aesthetic Quality in Text-to-Image Generation](https://mlsanity.com/paper/2402.17245/) (2024)
- [Scaling Laws Are Unreliable for Downstream Tasks: A Reality Check](https://mlsanity.com/paper/2507.00885/) (2025)
- [Too Large; Data Reduction for Vision-Language Pre-Training](https://mlsanity.com/paper/2305.20087/) (2023)
- [Performance Scaling via Optimal Transport: Enabling Data Selection from Partially Revealed Sources](https://mlsanity.com/paper/2307.02460/) (2023)
- [Optimizing Data Collection for Machine Learning](https://mlsanity.com/paper/2210.01234/) (2022)
- [Data Scaling Laws for End-to-End Autonomous Driving](https://mlsanity.com/paper/2504.04338/) (2025)
- [Delegated Classification](https://mlsanity.com/paper/2306.11475/) (2023)
- [Full or Weak annotations? An adaptive strategy for budget-constrained annotation campaigns](https://mlsanity.com/paper/2303.11678/) (2023)
- [Breaking Neural Network Scaling Laws with Modularity](https://mlsanity.com/paper/2409.05780/) (2024)
- [A Meta-Learning Approach to Predicting Performance and Data Requirements](https://mlsanity.com/paper/2303.01598/) (2023)
- [Unified Neural Network Scaling Laws and Scale-time Equivalence](https://mlsanity.com/paper/2409.05782/) (2024)
- [Bridging Data Gaps of Rare Conditions in ICU: A Multi-Disease Adaptation Approach for Clinical Prediction](https://mlsanity.com/paper/2507.06432/) (2025)
- [CopyScope: Model-level Copyright Infringement Quantification in the Diffusion Workflow](https://mlsanity.com/paper/2311.12847/) (2023)
- [A Probabilistic Method to Predict Classifier Accuracy on Larger Datasets given Small Pilot Data](https://mlsanity.com/paper/2311.18025/) (2023)
- [AdaDeDup: Adaptive Hybrid Data Pruning for Efficient Large-Scale Object Detection Training](https://mlsanity.com/paper/2507.00049/) (2025)
- [NRSeg: Noise-Resilient Learning for BEV Semantic Segmentation via Driving World Models](https://mlsanity.com/paper/2507.04002/) (2025)
- [Small-Scale Experiments: Are We There Yet?](https://mlsanity.com/paper/2608.11859/) (2026)
- [How much data do you need? Part 2: Predicting DL class specific training dataset sizes](https://mlsanity.com/paper/2403.06311/) (2024)
- [ViCTr: Vital Consistency Transfer for Pathology Aware Image Synthesis](https://mlsanity.com/paper/2505.04963/) (2025)
- [LCDB 1.1: A Database Illustrating Learning Curves Are More Ill-Behaved Than Previously Thought](https://mlsanity.com/paper/2505.15657/) (2025)
- [Strategies and impact of learning curve estimation for CNN-based image classification](https://mlsanity.com/paper/2310.08470/) (2023)
- [How Much Data are Enough? Investigating Dataset Requirements for Patch-Based Brain MRI Segmentation Tasks](https://mlsanity.com/paper/2404.03451/) (2024)
- [Target-Constrained Pooling for Trajectory Extrapolation](https://mlsanity.com/paper/2610.13034/) (2026)

## Abstract (arXiv)

Given a small training data set and a learning algorithm, how much more data is necessary to reach a target validation or test performance? This question is of critical importance in applications such as autonomous driving or medical imaging where collecting data is expensive and time-consuming. Overestimating or underestimating data requirements incurs substantial costs that could be avoided with an adequate budget. Prior work on neural scaling laws suggest that the power-law function can fit the validation performance curve and extrapolate it to larger data set sizes. We find that this does not immediately translate to the more difficult downstream task of estimating the required data set size to meet a target performance. In this work, we consider a broad class of computer vision tasks and systematically investigate a family of functions that generalize the power-law function to allow for better estimation of data requirements. Finally, we show that incorporating a tuned correction factor and collecting over multiple rounds significantly improves the performance of the data estimators. Using our guidelines, practitioners can accurately estimate data requirements of machine learning systems to gain savings in both development time and data acquisition costs.