Research
On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,695 papers · 148 categories

Trend · papers per month

245491736981 · Jun 202019922001200920172026
48 results for training rate

Learning-rate schedules for large models match optimization theory closely, leading to better training.

problem Improving training of large models with optimal learning rates.
method Used a bound from non-smooth convex optimization theory to match learning-rate schedules with practical benefits.
result Extending the learning-rate schedule with optimal learning-rate and transferring it across schedules improves model training.

WSqD extends learning rate schedules for large model training without fixed horizons.

problem Fixed learning rate schedules limit training horizon extension.
method WSqD replaces constant stable phase with a shifted inverse-square-root base, retaining linear cooldown.
result WSqD achieves minimax-optimal convergence rate and horizon-independence.

Large learning rates improve neural network generalization, study shows.

problem Understanding why large learning rates lead to better neural network generalization.
method Visual analysis of training and testing loss landscapes, introduction of a nonlinear model.
result Extended phase with large learning rates leads to near-optimal generalization.

We improve private training accuracy with learning rate schedules and matrix factorizations.

problem Private training with learning rate schedules and correlated noise.
method General upper and lower bounds for learning rate schedules, memory-efficient constructions, and schedule-aware factorizations.
result Schedule-aware factorizations improve accuracy in private training.

The learning rate is one of the most important hyper-parameters for model training and generalization. However, current hand-designed parametric learning rate schedules offer limited flexibility and the predefined schedule may not match the training dynamics of high dimensional and non-convex optimization problems. In …

2019-09-20abs ↗pdf ↗

GOLS-I automatically determines learning rates for various neural network training algorithms.

problem Adapting learning rates in stochastic training algorithms for neural networks.
method Gradient-Only Line Search (GOLS-I) for automatically setting learning rates.
result GOLS-I learning rate schedules are competitive with manually tuned rates across multiple algorithms, architectures, datasets, and loss functions.

Large learning rates cause oscillations in NN weights that improve generalization.

problem Improving generalization of neural networks trained with large learning rates.
method Theoretical analysis and feature-noise data generation model.
result Oscillating SGD with large learning rates benefits NN generalization by effectively learning weak features.

Maximal initial learning rate for deep ReLU networks identified.

problem Finding the optimal initial learning rate for deep neural networks.
method Simple approach to estimate maximal initial learning rate ηη^{\ast}, analyzing its behavior in constant-width fully-connected ReLU networks.
result Maximal initial learning rate ηη^{\ast} is well predicted as a power of depth × width, with specific conditions for network width and input layer training.

AdaScale SGD adapts learning rates for large-batch training efficiently.

problem Adapting learning rates for large-batch training to balance speed-ups and model quality.
method Adaptive learning rate adaptation based on gradient variance.
result AdaScale achieves reliable speed-ups for a wide range of batch sizes without degrading model quality.

Single training run learns optimal VAE parameters for various β values.

problem Training VAEs with varying β values for optimal trade-off between distortion and rate.
method Introduced Multi-Rate VAE (MR-VAE) using hypernetworks to map β to optimal parameters.
result MR-VAEs can construct the full rate-distortion curve without additional training.

Empirical study on SGD hyperparameters and adversarial robustness.

problem Effect of SGD hyperparameters on adversarial robustness and generalization.
method Empirical observation of learning rate, batch size, and momentum effects on adversarial robustness and generalization.
result Constant learning rate to batch size ratio leads to good generalization and almost constant adversarial robustness.

FIRE PBT improves neural network training by focusing on long-term performance.

problem Greedy decision mechanisms in PBT lead to poor long-term performance.
method FIRE PBT uses a fitness metric to encourage long-term performance over short-term improvements.
result FIRE PBT outperforms PBT on ImageNet and matches hand-tuned learning rates.

Gradient descent trains neural networks to match kernel regression's sharp generalization rate.

problem Training over-parameterized neural networks for nonparametric regression.
method Gradient descent with early stopping on over-parameterized two-layer neural networks.
result Trained neural networks achieve sharp generalization rate of O(εn2)\mathcal{O}(ε_n^2).

ScheduleFree+ improves large language model training without schedules or learning rates.

problem Scaling up Schedule-Free Learning to large language models.
method Learning-rate-free and schedule-free method for training large language models.
result ScheduleFree+ outperforms SOTA schedules by 31% at 1000 tokens per parameter.

Warm-up improves training by adapting learning rate based on curvature.

problem Improving training efficiency in deep learning models.
method Introducing a curvature condition to explain warm-up, and showing empirically that it leads to faster convergence.
result Adapting learning rate based on curvature condition naturally induces warm-up-like schedule, leading to faster convergence.

Adaptive weights improve physics-informed neural networks and deep operator networks.

problem Training physics-informed neural networks and deep operator networks can be challenging, leading to unsatisfactory accuracy and efficiency.
method Proposes a pointwise adaptive weighting method that balances the residual decay rate across different training points.
result Our proposed approach of balanced residual decay rates offers advantages including bounded weights, high prediction accuracy, fast convergence rate, low training uncertainty, low computational cost, and ease of hyperparameter tuning.

Learning rate decay (lrDecay) is a \emph{de facto} technique for training modern neural networks. It starts with a large learning rate and then decays it multiple times. It is empirically observed to help both optimization and generalization. Common beliefs in how lrDecay works come from the optimization analysis of (S…

2019-08-05abs ↗pdf ↗

Training deep neural networks using a large batch size has shown promising results and benefits many real-world applications. However, the optimizer converges slowly at early epochs and there is a gap between large-batch deep learning optimization heuristics and theoretical underpinnings. In this paper, we propose a no…

2020-02-04abs ↗pdf ↗

Most optimizers including stochastic gradient descent (SGD) and its adaptive gradient derivatives face the same problem where an effective learning rate during the training is vastly different. A learning rate scheduling, mostly tuned by hand, is usually employed in practice. In this paper, we propose CProp, a gradient…

2019-12-24abs ↗pdf ↗

Develops methods to measure and set function-space learning rates in neural networks.

problem Measuring and optimizing changes in neural network output functions.
method Efficient methods to measure and set function-space learning rates, requiring minimal computational overhead.
result Demonstrates FLeRM (Function-space Learning Rate Matching) for hyperparameter transfer across model scales.

Training deep neural networks requires intricate initialization and careful selection of learning rates. The emergence of stochastic gradient optimization methods that use adaptive learning rates based on squared past gradients, e.g., AdaGrad, AdaDelta, and Adam, eases the job slightly. However, such methods have also …

2019-10-27abs ↗pdf ↗

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.

This paper demonstrates dynamic hyper-parameter setting, for deep neural network training, using Mutual Information (MI). The specific hyper-parameter studied in this paper is the learning rate. MI between the output layer and true outcomes is used to dynamically set the learning rate of the network through the trainin…

2018-05-18abs ↗pdf ↗

New insights into how large learning rates affect transformer training dynamics.

problem Understanding how large learning rates impact the training of transformer models.
method Analyzing a simplified linear transformer model with a two-factor product map.
result Large learning rates can lead to various training outcomes including cycles, chaos, or divergence.

We explore a new way to evaluate generative models using insights from evaluation of competitive games between human players. We show experimentally that tournaments between generators and discriminators provide an effective way to evaluate generative models. We introduce two methods for summarizing tournament outcomes…

2018-08-14abs ↗pdf ↗

GD-trained shallow ReLU nets learn Lipschitz functions with noise.

problem Learning Lipschitz functions with additive noise in overparameterized neural networks.
method Gradient Descent (GD) with early stopping, focusing on the Neural Tangent Kernel (NTK).
result Early-stopped GD achieves minimax optimal rates for learning Lipschitz functions.

The study introduces anytime learning schedules for large language models without fixed horizons.

problem Training large language models without knowing the total training horizon.
method Theoretical analysis and weight averaging to create anytime learning schedules.
result Theoretical and empirical evidence shows that weight averaging with simple step sizes can achieve comparable final loss to well-tuned cosine schedules.

Large learning rates improve generalization, but optimal ranges are narrower than previously thought.

problem Optimizing learning rates for neural network training.
method Detailed exploration of learning rate ranges in a simplified setup, validating findings in a practical setting.
result Optimal learning rate ranges are significantly narrower than previously assumed.

Model proposes neural network for continuous time dynamics with inductive biases.

problem Training neural networks for small datasets with nonlinear dynamics.
method Inductive biases on decay rates and frequencies using Koopman operator theory.
result Higher forecasting performance with single short training sequence.

It is common practice to decay the learning rate. Here we show one can usually obtain the same learning curve on both training and test sets by instead increasing the batch size during training. This procedure is successful for stochastic gradient descent (SGD), SGD with momentum, Nesterov momentum, and Adam. It reache…

2017-11-01abs ↗pdf ↗

Paper finds wide minima are better for generalization and proposes a new learning rate schedule.

problem The challenge of finding optimal learning rates for model training.
method The paper introduces a new hypothesis about the density of wide minima and designs an explore-exploit learning rate schedule.
result The explore-exploit learning rate schedule improves model performance and reduces training time.

A hybrid ASR system using conformer architecture improves word-error-rate and training speed.

problem Improving word-error-rate and training efficiency for hybrid ASR systems.
method Used conformer architecture, applied time downsampling, and transposed convolutions.
result Conformer-based hybrid model achieves competitive results and significantly outperforms BLSTM-based hybrid model.