Preprocessing data is an important step before any data analysis. In this paper, we focus on one particular aspect, namely scaling or normalization. We analyze various scaling methods in common use and study their effects on different statistical learning models. We will propose a new two-stage scaling method. First, w…
This paper improves generative models by using data scaling and theoretical analysis.
problem Challenges in selecting noise distributions for stable learning in generative models.
method Introduces Scale-GAN, which uses data scaling and variance-based regularization.
result Data scaling controls the bias-variance trade-off and improves stability and accuracy.
We predict generalization error across model and dataset sizes.
problem Understanding the dependency of neural network generalization error on model and dataset size.
method Model scaling concept applied to construct a functional form of generalization error.
result The constructed functional form accurately predicts generalization error across scales.
This paper reviews some of the phenomenological models which have been introduced to incorporate the scaling properties of financial data. It also illustrates a microscopic model, based on heterogeneous interacting agents, which provides a possible explanation for the complex dynamics of markets' returns. Scaling and m…
Study presents MMC model for better fitting multiple choice data.
problem Improving accuracy of latent trait estimates in IRT models.
method Fit autoencoders to MMC model, demonstrating better fit than nominal response model.
result MMC model outperforms traditional IRT models in fit.
New phases identified in neural scaling laws with compute limits.
problem Understanding neural scaling laws under compute constraints.
method Solved neural scaling model with stochastic gradient descent, derived loss curves, analyzed model-parameter-count phases.
result Identified 4 phases (+3 subphases) in data-complexity/target-complexity phase-plane, derived exponents.
Unified scaling laws reveal how model size and training time impact neural network performance.
problem Understanding how much performance improvement can be expected from scaling model size or data volume.
method Established scale-time equivalence and combined it with a linear model analysis of double descent.
result Unified theoretical scaling laws explain previously unexplained phenomena and offer a more accessible path to training large models.
Neural networks' performance scales with data size, explained by data manifold dimensionality.
problem Understanding the scaling of neural network performance with the number of parameters.
method Explained by the intrinsic dimension of the data manifold, confirmed through teacher/student framework and various datasets.
result The scaling exponent α is approximately 4 divided by the intrinsic dimension d of the data manifold.
This work explores variably scaled kernels to improve non-stationary Gaussian processes.
problem Limited ability of stationary kernels to represent heterogeneous correlation structures.
method Introduces variably scaled kernels to modify correlation structures explicitly.
result Improved reconstruction accuracy and better uncertainty estimates for non-stationary data.
Scaling laws in linear regression explain model performance improvements with size and data.
problem Disagreement between empirical neural scaling laws and conventional wisdom on variance error.
method Infinite dimensional linear regression setup, one-pass SGD, Gaussian prior, power-law spectrum.
result Variance error is dominated by other errors, disappearing from the bound due to SGD's implicit regularization.
Improved loss scaling for stochastic momentum algorithms in high dimensions.
problem Improving loss scaling for stochastic momentum algorithms in high dimensions.
method Dimension-adapted Nesterov acceleration (DANA) scales momentum hyperparameters based on model size and data complexity.
result DANA improves loss scaling exponents across various data and target complexities.
We determine the critical batch size for large language models and find it scales with data size, not model size.
problem Determining the optimal batch size for large-scale model training.
method We propose a measure of critical batch size, pre-trained models, and systematic hyper-parameter sweeps.
result The critical batch size scales primarily with data size, not model size.
AutoScale improves LLM pre-training by adjusting data mixtures at different scales.
problem Data mixtures that work well at small scales may not perform as well at larger scales.
method AutoScale uses a two-stage approach: fitting a model to predict loss under different compositions and extrapolating optimal compositions to larger scales.
result AutoScale accelerates convergence and improves downstream performance.
Scaling laws govern predictive uncertainties in deep learning models.
problem Understanding predictive uncertainties in deep learning models.
method Empirical analysis and approximate Bayesian inference on vision and language tasks.
result Scaling laws exist for various measures of predictive uncertainty in deep learning models.
Study how generalization scales with model size and data in quadratic neural networks.
problem Understanding how generalization scales with model size and data in quadratic neural networks.
method Analyzed ℓ2-regularized empirical test error minimization in a quadratic two-layer network with finite-sample setting and structured data. result Revealed a phase diagram with distinct scaling regimes as the number of parameters varies, showing data-dependent power laws controlled by spectral structure of the target.
The study explains transformer scaling laws using statistical and approximation theories.
problem Understanding why transformer scaling laws exist for large models trained on low-dimensional data.
method Established statistical estimation and mathematical approximation theories for transformers on low-dimensional manifolds.
result Predicted a power law between generalization error and model and data sizes, with power depending on intrinsic data dimension.
Study shows cliff-learning in transfer learning from foundation models.
problem Data-scaling of transfer learning from foundation models in low data regimes.
method Investigation of cliff-learning phenomenon through foundation-model analysis and toy models.
result Cliff-learning reflects compatibility between priors and tasks.
In this paper, we consider daily financial data of a collection of different stock market indices, exchange rates, and interest rates, and we analyze their multi-scaling properties by estimating a simple specification of the Markov-switching multifractal model (MSM). In order to see how well the estimated models captur…
Theory explains neural network scaling with dataset and model size.
problem Neural network scaling laws with dataset and model size.
method Identified variance-limited and resolution-limited scaling behaviors.
result Four scaling regimes explained: infinite data, infinite width, resolution-limited, and large width.
New kernel model scales to large datasets.
problem Challenges in scaling kernel machines to large datasets.
method Introduced EigenPro 3.0, an algorithm based on projected dual preconditioned SGD.
result Ability to scale model and data sizes independently.
CrossAD detects anomalies in time series data by considering cross-scale associations and cross-window modeling.
problem Anomaly detection in time series data is challenging due to varying patterns at different scales and fixed window sizes.
method CrossAD incorporates cross-scale reconstruction and a query library to capture dynamic cross-scale associations and comprehensive context.
result CrossAD achieves state-of-the-art performance in anomaly detection across multiple real-world datasets.
Improved scaling laws in linear regression using data reuse.
problem Sustainability of neural scaling laws when running out of new data.
method Data reuse in multi-pass stochastic gradient descent (multi-pass SGD) for M-dimensional linear models trained on N data with sketched features. result Multi-pass SGD achieves a test error of Θ(M1−b+L(1−b)/a) with L>N, improving scaling laws in data-constrained regimes. Sloth predicts LLM performance using latent skills across families.
problem Variations in benchmark performance due to differences in training configurations and data processing across model families.
method Sloth uses publicly available benchmark data and assumes LLM performance is driven by latent skills influenced by model size and training tokens. It exploits correlations across benchmarks to provide accurate predictions.
result Sloth predicts LLM performance accurately and offers insights into scaling behaviors for complex tasks.
Deep learning (DL) creates impactful advances following a virtuous recipe: model architecture search, creating large training data sets, and scaling computation. It is widely believed that growing training sets and models should improve accuracy and result in better products. As DL application domains grow, we would li…
Study uses CNNs to upscale wind speed data from 100 km to 3 km, improving subgrid-scale variability.
problem Recovering fine-scale wind speed information from coarse data.
method Convolutional neural networks (CNNs) with different input configurations (coarse wind speed, fine-scale topography, diurnal cycle) were tested.
result CNN models with coarse wind and fine topography inputs perform best in generalizing to unseen regions.
Study reveals neural scaling laws in random graphs and natural language models.
problem Understanding the origin of neural scaling laws in complex systems.
method Examined scaling laws in transformers trained on random walks and simplified natural language models.
result Neural scaling laws emerge in the absence of power law structure in data correlations.
Early fault detection using instrumented sensor data is one of the promising application areas of machine learning in industrial facilities. However, it is difficult to improve the generalization performance of the trained fault-detection model because of the complex system configuration in the target diagnostic system…
This work studies scaling laws for low-precision training in high-dimensional linear regression.
problem Optimizing trade-off between model quality and training costs in high-dimensional linear regression.
method Theoretical study of scaling laws for low-precision training within a high-dimensional sketched linear regression framework, analyzing multiplicative and additive quantization.
result Multiplicative quantization maintains full-precision model size, while additive quantization reduces effective model size.
Transformers can scale both context and task, but MLPs can only scale task.
problem Understanding and scaling In-Context Learning in transformers.
method Simplified transformer architecture, feature map, and MLP combination.
result Simplified transformer can perform ICL and context-scaling but not task-scaling.
Adding data can sometimes hurt model performance in multi-source healthcare tasks.
problem Identifying when adding more data helps or hinders model outcomes in multi-source healthcare tasks.
method Identified the Data Addition Dilemma, demonstrated empirically observed trade-offs, introduced distribution shift heuristics.
result Adding data can sometimes reduce model performance due to distribution shift.
Theory predicts neural scaling exponents from language statistics.
problem No existing theory could quantitatively predict neural scaling exponents.
method Isolated two key statistical properties of language.
result Derives a simple formula predicting neural scaling exponents.
DD-SP uses ML to improve SP for Lorenz 96 systems, outperforming LR and DD-P.
problem Improving computational efficiency in weather/climate modeling.
method Data-driven super-parameterization using recurrent neural networks.
result DD-SP is more accurate and cheaper than SP, especially with scale separation.
Proposes a method to predict time series data using scale information.
problem Sequential data exhibits different behavior at different scales.
method Scale-based inference using k-nearest neighbors and sequential prediction. result Asymptotically, the algorithm produces the best prediction error.
FaStR improves scalability for time-aware RS with varying coefficients.
problem Limited applicability of structured regression models to large-scale data with categorical effects and many interactions.
method Combines structured additive regression and factorization approaches in a neural network-based model implementation.
result FaStR scales better and performs competitively with other time-aware RS in prediction performance.
The paper predicts loss scaling across different datasets and compute scales.
problem Predicting loss scaling across different datasets and compute scales.
method Derive shifted power law relationships between train and test losses.
result Shifted power law relationships hold for various datasets and tasks, improving prediction accuracy.
This work explains scaling laws as redundancy laws in deep learning.
problem The mathematical origins of scaling laws in deep learning models remain unclear.
method Kernel regression and analysis of data covariance spectra.
result Scaling laws can be explained as redundancy laws, revealing the learning curve's slope depends on data redundancy.
Develops a simple model to understand learning curves for arbitrary power laws.
problem Lack of theoretical understanding of scaling laws in machine learning.
method Analyzes a toy model to determine if learning curves are universal or depend on data distribution.
result Determines that learning curves can exhibit n−β for arbitrary power β>0. New pruning method breaks power law scaling, potentially reducing error to exponential.
problem Improving neural network performance through scaling alone is costly.
method Developed a new data pruning metric to break power law scaling.
result Pruned datasets show better than power law scaling on various image datasets.
Efficiently trains deep Gaussian processes on large datasets.
problem Large-scale data and multi-scale features in function approximation.
method Combines variational learning with MCMC for efficient and accurate training.
result Highly efficient and accurate deep GP training on large-scale data.
DRFormer uses dynamic tokenization and multi-scale transformer to forecast long time series.
problem Forecasting long-term time series data across diverse scales.
method Dynamic tokenizer, multi-scale transformer, dynamic sparse learning, rotary position encoding.
result DRFormer outperforms existing methods in forecasting accuracy.
Improved model accuracy can reduce overall user accuracy in competitive markets.
problem The impact of model competition on overall user accuracy.
method Defined a model of competition for classification tasks and used data representations to study the effect of scale.
result Improving data representation quality can decrease overall predictive accuracy across users (social welfare) in a competitive market.
Large models follow power laws in performance with dataset size or parameters.
problem Understanding neural scaling laws in large language models.
method Joint generative data model and random feature model.
result Modeling and solving the dual limit reveals insights into scaling laws.
Much recent work has concerned sparse approximations to speed up the Gaussian process regression from the unfavorable O(n3) scaling in computational time to O(nm2). Thus far, work has concentrated on models with one covariance function. However, in many practical situations additive models with multiple covariance func…
PePR scores assess DL model performance per resource unit, promoting smaller, more efficient models.
problem Limited access to large-scale resources hinders medical image analysis research.
method Introduced PePR score to measure DL model performance per resource unit.
result Small-scale, specialized models outperform large-scale models in resource-constrained settings.
The most common stochastic volatility models such as the Ornstein-Uhlenbeck (OU), the Heston, the exponential OU (ExpOU) and Hull-White models define volatility as a Markovian process. In this work we check of the applicability of the Markovian approximation at separate times scales and will try to answer the question …
A digital twin for multi-scale systems uses physics-based and machine learning models.
problem Lack of application-specific details in digital twin technology.
method Strategically separates into physics-based and data-driven models; uses mixture of experts with Gaussian Process.
result Robust and accurate predictions at future time-steps for multi-scale systems.
A new scaling law predicts optimal batch size for training models.
problem Finding the optimal batch size for training models efficiently.
method Proposed a three-term scaling law that considers model size, training data, training steps, and batch size.
result The three-term law accurately recovers the optimal batch size and can be robustly fit with fewer training runs.
Automates PDE model reduction with time-scale separation.
problem Computational expense in solving high-dimensional PDEs.
method Combines autoencoder and time-continuous model for latent dynamics.
result Automatically learns independent temporal scales in complex systems.