Ensembles of random-feature models can't outperform a single large model.
problem Finding the optimal balance between model size and ensemble size.
method Deterministic equivalent risk estimates and scaling laws analysis.
result Ensembles of random-feature models achieve near-optimal performance only under specific conditions.
Study shows how algorithmic choices affect optimal batch sizes in neural networks.
problem Understanding how batch size impacts neural network training efficiency.
method Experiments and analysis of a simple quadratic model to study algorithmic choices.
result Preconditioned optimizers like Adam and K-FAC allow larger batch sizes before diminishing returns.
Optimal microlending group size is 5 people.
problem Determining the best group size for microlending to minimize default risk.
method Mathematical modeling with interacting forces and precise hypotheses.
result The optimal microlending group size is 5 people.
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.
Scaling laws for neural language models reveal optimal model size and compute allocation.
problem Understanding the optimal model size and compute allocation for neural language models.
method Empirical analysis of scaling laws for cross-entropy loss across model size, dataset size, and compute.
result Simple equations govern the dependence of overfitting and training speed on model/dataset size and model size, respectively.
As neural networks become widely deployed in different applications and on different hardware, it has become increasingly important to optimize inference time and model size along with model accuracy. Most current techniques optimize model size, model accuracy and inference time in different stages, resulting in subopt…
We propose an adaptive optimization method for deep learning that dynamically adjusts batch size.
problem Optimizing deep learning models with varying sensitivity to batch size selection.
method Adaptive regularization with dynamically determined stochastic batch size based on gradient norms.
result Our method outperforms state-of-the-art optimization algorithms in generalization and robustness.
A simple block configures optimal kernel sizes for time series classification.
problem Choosing the right kernel size for time series classification.
method Proposes Omni-Scale block (OS-block) with kernel sizes determined by prime numbers.
result Models with OS-block achieve state-of-the-art performance on time series benchmarks.
Adaptive batch size schedules improve language model training efficiency and generalization.
problem Dilemma of choosing batch sizes in large-scale model training.
method General-purpose adaptive batch size schedules compatible with data and model parallelism.
result Adaptive batch size schedules outperform constant batch sizes and heuristic warmup schedules.
Seesaw optimizes training by balancing learning rate and batch size, accelerating model pretraining.
problem Optimizing training efficiency for large language models with adaptive optimizers.
method Develops a principled framework for batch-size scheduling, introducing Seesaw which multiplies learning rate by 1/√2 and doubles batch size.
result Empirically, Seesaw reduces wall-clock time by approximately 36% compared to cosine decay, matching theoretical limits.
AdAdaGrad optimizes batch sizes for deep learning models, reducing the generalization gap.
problem The generalization gap between large-batch and small-batch training in deep learning.
method AdAdaGrad introduces adaptive batch size strategies derived from adaptive sampling methods.
result AdAdaGradNorm converges to a first-order stationary point with a rate of O(1/K) in K iterations.
We present the Integrated Size and Price Optimization Problem (ISPO) for a fashion discounter with many branches. Based on a two-stage stochastic programming model with recourse, we develop an exact algorithm and a production-compliant heuristic that produces small optimality gaps. In a field study we show that a distr…
This work provides a scaling rule for model EMA optimization across batch sizes.
problem Training dynamics and performance differences across batch sizes when using model EMA.
method Developed a scaling rule for model EMA optimization, demonstrating its validity across various architectures and data modalities.
result Enabled SSL methods like BYOL to train at larger batch sizes without performance degradation.
NIS learns optimal embedding sizes for recommendation models.
problem Finding optimal embedding sizes for large-scale recommendation models.
method Neural Input Search (NIS) uses reinforcement learning to automatically find optimal vocabulary sizes and embedding dimensions.
result NIS improves prediction accuracy by 6.8% on Recall@1 and 1.8% on ROC-AUC.
Proposes an exponentially increasing step-size for faster parameter estimation in statistical models.
problem Slow convergence of gradient descent in locally convex loss functions.
method Exponentially increasing step-size in gradient descent algorithm.
result Converges linearly to optimal solution under homogeneous assumptions.
Optimal regularization can prevent the double descent phenomenon in learning models.
problem The double descent phenomenon in learning models, where test performance is non-monotonic in sample size and model size.
method Theoretical and empirical study of optimal ℓ 2 \ell_2 ℓ 2 regularization for linear regression models and neural networks. result Optimally-tuned ℓ 2 \ell_2 ℓ 2 regularization achieves monotonic test performance for certain models and mitigates the double descent phenomenon for more general models. New method improves model accuracy in Byzantine-robust distributed learning by optimizing batch size.
problem Reduces model accuracy drop due to large variance of stochastic gradients in Byzantine-robust distributed learning.
method Proposes ByzSGDnm, a novel BRDL method that uses normalized momentum to mitigate accuracy drop in large batch sizes.
result The optimal batch size increases with the fraction of Byzantine workers, leading to better model accuracy under Byzantine attacks.
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.
Optimal scaling found to depend on operator norm across large models and datasets.
problem Lack of unifying principle for optimal hyperparameter scaling across models and datasets.
method Discovered that optimal scaling is conditioned on the operator norm of the output layer.
result The optimal learning rate/batch size pair ( η ∗ , B ∗ ) (η^{\ast}, B^{\ast}) ( η ∗ , B ∗ ) consistently has the same operator norm value. New analysis shows optimal embedding learning rate depends on vocabulary size, not just model width.
problem Optimal learning rate for language model embeddings is not well understood, especially with large vocabularies.
method Theoretical analysis of training dynamics, interpolation between μ μ μ P and LV regimes. result Optimal embedding learning rate scales as Θ ( w i d t h ) Θ(\sqrt{width}) Θ ( w i d t h ) in the LV regime, not Θ ( w i d t h ) Θ(width) Θ ( w i d t h ) as μ μ μ P predicts. Study optimizes step size for Metropolis algorithm in non-identifiable cases.
problem Optimizing step size for Metropolis algorithm in non-identifiable models.
method Analytical derivation of average acceptance rate for non-identifiable cases.
result Developed optimization principle for step size based on average acceptance rate.
Muon optimizes training efficiency by improving data retention at large batch sizes.
problem Improving training efficiency and data retention at large batch sizes.
method Introducing Muon, a second-order optimizer, and combining it with muP for efficient hyperparameter transfer.
result Muon outperforms AdamW in retaining data efficiency at large batch sizes, enabling more economical training.
Developed AI models for multi-gas detection in near IR spectrums.
problem Detecting multiple gases in near IR spectrums.
method Used Monte Carlo KNN and multi-resolution CNN, synthesized near IR spectrums, optimized kernel sizes and channels.
result Multi-resolution CNN outperforms other models.
Paper overcomes sample size barrier in reinforcement learning with generative models.
problem Sample efficiency in reinforcement learning with generative models.
method Developed two algorithms to certify minimax optimality of sample complexity.
result Achieved minimax-optimal guarantees for a wide range of sample sizes.
This work improves SGD convergence by adaptively adjusting batch sizes.
problem High variance in gradient estimates with small batch sizes.
method Adaptive batch size adjustment based on model training loss.
result Adaptive batch size method requires fewer model updates with same computation.
This paper studies the Glosten Milgrom model whose risky asset value admits an arbitrary discrete distribution. Contrast to existing results on insider's models, the insider's optimal strategy in this model, if exists, is not of feedback type. Therefore a weak formulation of equilibrium is proposed. In this weak formul…
Adaptive SGD learns optimal batch size for strong convex functions.
problem Finding optimal batch size for SGD in practice.
method Adaptive SGD method that learns optimal batch size.
result Adaptive SGD exhibits nearly optimal performance in experiments.
Model predicts neural network performance scaling laws across various factors.
problem Understanding the performance of neural networks across different training factors.
method Random feature model trained with gradient descent, analyzing compute-optimal scaling laws.
result Predicts asymmetric compute-optimal scaling rule and behavior of training and test loss gap.
MBS reduces CNN model size with minimal accuracy loss.
problem Reducing CNN model size while maintaining accuracy.
method Adaptive macroblock scaling based on effective flops.
result Significant model size reduction across various CNN architectures.
Standard optimizers perform as well as LARS and LAMB at large batch sizes.
problem Comparing optimizers for neural network training at large batch sizes.
method Used standard optimizers like Nesterov momentum and Adam to match or exceed LARS and LAMB results.
result Standard optimizers can match or exceed LARS and LAMB at large batch sizes.
A new algorithm improves efficiency and robustness of heuristic optimization in simulation-based problems.
problem Optimizing input parameters for stochastic simulation-based optimization.
method Reactive sample size algorithm based on parametric tests and indifference-zone selection.
result The reactive method improves efficiency and robustness of heuristic optimization techniques.
This paper uses supervised learning to predict optimal chunk-size for parallel linear algebra operations.
problem Finding the optimal chunk-size for parallel linear algebra operations.
method The paper uses supervised learning models (logistic regression, neural networks, decision trees) to predict the optimal chunk-size for multiple linear algebra operations.
result The custom decision tree model outperforms classical decision trees and other models in predicting optimal chunk-size for linear algebra operations.
SGD's performance improves with critical batch size, minimizing SFO complexity.
problem Optimizing SGD's performance with batch size and learning rate.
method Analysis of SGD using constant and decaying learning rates, focusing on batch size effects.
result SGD with critical batch size minimizes SFO complexity.
New algorithm improves stability of optimization algorithms by adapting step-size.
problem Optimization algorithms' effectiveness is sensitive to step-size hyperparameters.
method Adapts NGN step-size method with momentum to enhance stability.
result Achieves convergence rate of O(1/√K) without restrictive assumptions.
Improved variance reduction for Riemannian non-convex optimization with adaptive batch size.
problem Optimizing non-convex functions on Riemannian manifolds.
method Batch size adaptation in R-SVRG, R-SRG, and R-SPIDER.
result Achieves lower total complexities for various non-convex functions.
We make policy optimization algorithms batch size-invariant by decoupling proximal and behavior policies.
problem Some policy optimization algorithms do not have batch size-invariance, leading to inefficiencies.
method We decouple the proximal policy from the behavior policy to achieve batch size-invariance.
result Our approach makes policy optimization algorithms more efficient and allows them to use stale data more effectively.
Technique creates highly accurate small models for better interpretability.
problem Balancing model accuracy and interpretability for constrained models.
method Identifies optimal training distribution for a given model size using Infinite Mixture Model with Beta components and Bayesian Optimization.
result Significant improvements in F1-score, up to 100% in some cases.
Downsampling can improve generalization in ridgeless linear regression, especially with optimal sketching size.
problem Improving generalization in ridgeless linear regression with limited data.
method Investigating the effects of downsampling on the sketched ridgeless least square estimator in the proportional regime.
result Optimal sketching size minimizes out-of-sample prediction risks and stabilizes risk curves.
Worst-Case Sensitivity measures model sensitivity to uncertainty set size.
problem Model sensitivity to uncertainty set size in Distributionally Robust Optimization.
method Introducing Worst-Case Sensitivity as a measure of model sensitivity, and deriving closed-form expressions for various uncertainty sets.
result DRO solutions can be sensitive to the family and size of the uncertainty set, and worst-case sensitivity reflects these properties.
New framework optimizes deep learning training by deferring large batch sizes to late stages.
problem Optimizing batch size scheduling for deep learning training efficiency.
method Introduced the functional scaling law (FSL) framework to analyze and optimize batch size scheduling.
result Large batch sizes can be deferred to late training stages without sacrificing performance.
This paper tackles multi-asset market making by reducing dimensionality and considering different transaction sizes.
problem Optimizing bid and ask prices for multiple assets while managing inventory risk in volatile markets.
method Proposes a dimensionality reduction technique using a factor model and considers different transaction sizes.
result Generalizes existing market making models by incorporating different transaction sizes and prices.
Investment strategy depends on many factors for venture capital funds.
problem Finding the optimal portfolio size for venture capital funds.
method Analyzes various factors affecting fund returns and optimal portfolio size, starting with basic assumptions and increasing complexity.
result Investment strategy depends on many factors, not a one-size-fits-all formula.
Paper proposes optimal investment and reinsurance strategies considering financial and insurance risks dependence.
problem Optimal investment and reinsurance strategies under dependent financial and insurance risks.
method Stochastic control approach to maximize expected exponential utility of terminal wealth.
result Minimal dependence between financial and insurance risks significantly impacts investment and reinsurance strategies.
Improved SVI with adjustable annealing for better optimization.
problem Improving optimization in stochastic variational inference.
method Tuneable stochastic annealing in SVI with adjustable batch size.
result Approximation to maximum entropy stochastic gradient at desired variance level.
This work explores representation complexity in RL paradigms, revealing model-based RL as the easiest task.
problem Investigating the representation complexity gap among model-based, policy-based, and value-based RL.
method Demonstrated through analysis of Markov decision processes (MDPs) and introduced new classes of MDPs.
result Representation complexity hierarchy: model-based RL > policy-based RL > value-based RL.
New scaling laws optimize model size, training, and inference for better performance.
problem Trade-off between model size and inference cost in modern LLMs.
method Train-to-Test ( T 2 T^2 T 2 ) scaling laws that jointly optimize model size, training tokens, and inference samples. result Optimal pretraining decisions shift into overtraining regime, leading to stronger performance.
Paper proposes a classifier that optimizes utility function with prior knowledge.
problem Designing a classifier that optimizes a utility function based on prior knowledge.
method Systematic framework incorporating prior knowledge to optimize a utility function.
result The classifier asymptotically converges to the optimal classifier (Bayes rule) as data size grows.
Adaptive step-size improves optimization in complex geometries.
problem Optimizing functions with non-Euclidean geometries.
method Adaptive step-size strategy for optimization algorithms.
result Guaranteed convergence for Adaptive Conditional Gradient Descent.