Unintended effects from scaling neural network outputs with adaptive learning rates.
problem Adaptive learning rate optimization's behavior is altered by output scaling, leading to misinterpretation.
method Presented a modified optimization algorithm to mitigate unintended effects.
result Adaptive learning rate's effectiveness is significantly impacted by output scaling, especially for small scaling factors.
Study uses ML to analyze how interest rates affect fund returns, finding gradient boosting is effective.
problem Understanding how interest rate changes impact fund returns.
method Combines Machine Learning and causal inference, using Double Machine Learning framework.
result Gradient boosting is useful for predicting fund returns, showing a significant negative effect of interest rate increases.
The paper analyzes how learning rate affects SGD and provides insights into optimal rates.
problem Understanding the impact of learning rate on stochastic gradient descent.
method Developed a learning-rate-dependent stochastic differential equation (lr-dependent SDE) to analyze SGD.
result Established a linear rate of convergence for SGD and found the optimal linear rate by analyzing the spectrum of the Witten-Laplacian.
This paper investigates the nonparametric regression problem using SVMs with anisotropic Gaussian RBF kernels. Under the assumption that the target functions are resided in certain anisotropic Besov spaces, we establish the almost optimal learning rates, more precisely, optimal up to some logarithmic factor, presented …
Derives effective continuous dynamics for adaptive SGD methods.
problem Analyzing noise in adaptive SGD methods.
method Stochastic modified equations framework and Malladi's scaling rules.
result Sampling-induced noise in SGD limits to independent Brownian motions.
LALR adapts learning rate for faster convergence in regression and neural nets.
problem Finding optimal learning rates for faster convergence in regression and neural networks.
method Lipschitz continuity theory applied to Mean Absolute Error and Quantile loss functions.
result Adaptive learning rate policy enables up to 20x faster convergence.
Study separates learning rate effects from adaptive gradient methods.
problem Understanding the impact of learning rates on neural network training.
method Introduced a 'grafting' experiment to isolate learning rate effects.
result Many existing beliefs about adaptive gradient methods may be incorrect.
Adaptive learning rate algorithms such as RMSProp are widely used for training deep neural networks. RMSProp offers efficient training since it uses first order gradients to approximate Hessian-based preconditioning. However, since the first order gradients include noise caused by stochastic optimization, the approxima…
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.
SALR improves deep learning generalization by dynamically adjusting learning rates.
problem Improving generalization in deep learning models.
method Sharpness-aware learning rate scheduling based on local loss function sharpness.
result SALR drives solutions to flatter regions, improving generalization and convergence.
Study shows deep linear networks can converge to flatter minima at large learning rates.
problem Understanding the implicit bias of deep linear networks at large learning rates.
method Characterization of deep linear networks for binary classification using logistic loss in the large learning rate regime.
result Gradient descent iterates converge to a flatter minimum in the catapult phase for certain data separation conditions.
A new method to improve deep neural networks using weight rescaling.
problem Overfitting and sensitivity to hyperparameters in weight decay.
method Weight rescaling (WRS) to control weight norm and prevent overfitting.
result WRS outperforms weight decay and other methods in various applications.
This paper investigates robust and efficient DR/RDR estimators for WATEs.
problem Lack of systematic investigation into robustness and efficiency conditions for WATE estimation.
method Proposes three RDR estimators using semiparametric efficient influence function and double/debiased machine learning.
result Demonstrates the practical relevance of the methods in medical and social sciences.
The performance of stochastic gradient descent (SGD) depends critically on how learning rates are tuned and decreased over time. We propose a method to automatically adjust multiple learning rates so as to minimize the expected error at any one time. The method relies on local gradient variations across samples. In our…
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…
Study analyzes how discounts affect train ticket purchases and rescheduling in Switzerland.
problem Understanding how discounts influence train ticket buying and rescheduling behavior.
method Machine learning techniques, including causal machine learning, to analyze survey data.
result Increasing a discount rate by 1% increases the rescheduled trip share by 0.16% among always buyers.
We analyse the effectiveness of modern deep learning techniques in predicting credit ratings over a universe of thousands of global corporate entities obligations when compared to most popular, traditional machine-learning approaches such as linear models and tree-based classifiers. Our results show a adequate accuracy…
Random learning rate improves neural network training without extra cost.
problem Optimizing neural networks with stochastic gradient descent.
method Uniformly-distributed random learning rate in SGD.
result Random learning rate yields better regularization without extra computational cost.
Muon optimizer improves federated learning performance.
problem Improving federated learning performance with Muon optimizer.
method Proposed FedMuon algorithm with orthonormalized update direction.
result FedMuon's learning rate is independent of problem-specific parameters.
This paper explains why Adam generalizes worse than SGD by analyzing its components.
problem Understanding why Adam generalizes worse than Stochastic Gradient Descent (SGD).
method Diffusion theoretical framework to disentangle the effects of Adaptive Learning Rate and Momentum.
result Adaptive Learning Rate helps escape saddle points but not select flat minima, while Momentum provides a drift effect to help pass through saddle points.
This paper introduces an innovative Bayesian machine learning algorithm to draw interpretable inference on heterogeneous causal effects in the presence of imperfect compliance (e.g., under an irregular assignment mechanism). We show, through Monte Carlo simulations, that the proposed Bayesian Causal Forest with Instrum…
Large learning rates work surprisingly well in standard parameterization, contrary to theory.
problem Theoretical limits of large learning rates do not match practical network behavior.
method Fine-grained analysis of learning rates and network behavior under cross-entropy loss.
result There are two distinct sub-regimes of unstable learning rates, with a controlled divergence regime where features continue to evolve.
Large learning rates enhance model robustness and compressibility.
problem Achieving robustness and resource-efficiency in machine learning models.
method Identifying and utilizing large learning rates as a facilitator for robustness and compressibility.
result Large learning rates produce desirable representation properties and compare favorably to other methods.
New model explains deep learning performance at large learning rates.
problem Understanding deep learning performance at different learning rates.
method Developed neural networks with solvable training dynamics.
result Large learning rates lead to convergence to flatter minima.
Stochastic gradient descent with a large initial learning rate is widely used for training modern neural net architectures. Although a small initial learning rate allows for faster training and better test performance initially, the large learning rate achieves better generalization soon after the learning rate is anne…
Learning Rate (LR) is an important hyper-parameter to tune for effective training of deep neural networks (DNNs). Even for the baseline of a constant learning rate, it is non-trivial to choose a good constant value for training a DNN. Dynamic learning rates involve multi-step tuning of LR values at various stages of th…
Simpler, parameter-free AdaGrad and Adam variants with convergence guarantees.
problem Inefficiencies in ad-hoc learning rate tuning for optimization algorithms.
method Developed AdaGrad++ and Adam++ without predefined learning rates and proved their convergence.
result AdaGrad++ and Adam++ achieve comparable convergence rates to AdaGrad and Adam respectively.
Simplifies VAE for anomaly detection using rate-distortion theory.
problem Anomaly detection in unsupervised learning systems.
method Revisit VAE from information theory, incorporate model uncertainty.
result Competitive performance on benchmark datasets.
Recent work has established an empirically successful framework for adapting learning rates for stochastic gradient descent (SGD). This effectively removes all needs for tuning, while automatically reducing learning rates over time on stationary problems, and permitting learning rates to grow appropriately in non-stati…
ISOKANN learns collective variables and effective dynamics for metastable transitions.
problem Understanding metastable transitions in complex molecular systems.
method Integrates Koopman operators with neural networks to extract CVs and effective dynamics.
result Reconstructs coarse-grained kinetics and reproduces transition times across barriers.
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…
Privacy affects how much data is needed for CVaR optimization.
problem Privacy constraints impact the effective sample size for CVaR optimization.
method Analyzes the privacy-relevant sample size and decomposes CVaR excess risk.
result The effective private tail sample size is εnτ, affecting CVaR learning rates.
Active learning can't improve over passive in certain settings.
problem Active learning vs. passive learning in nonparametric settings.
method Analyzing margin conditions and their effects on active learning performance.
result Nuances in margin conditions determine whether active learning can outperform passive learning.
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 η∗, analyzing its behavior in constant-width fully-connected ReLU networks. result Maximal initial learning rate η∗ is well predicted as a power of depth × width, with specific conditions for network width and input layer training. Batch Normalization (BN) has become a cornerstone of deep learning across diverse architectures, appearing to help optimization as well as generalization. While the idea makes intuitive sense, theoretical analysis of its effectiveness has been lacking. Here theoretical support is provided for one of its conjectured pro…
We introduce a general method for improving the convergence rate of gradient-based optimizers that is easy to implement and works well in practice. We demonstrate the effectiveness of the method in a range of optimization problems by applying it to stochastic gradient descent, stochastic gradient descent with Nesterov …
Bayesian Optimization improves machine learning for detecting network attacks.
problem Detecting network attacks to secure critical information.
method Bayesian Optimization for tuning machine learning algorithms (SVM-RBF, RF, k-NN).
result The proposed framework achieves high accuracy and low false alarms.
Study finds flipped classrooms improve student self-concept, enjoyment, but not exam scores.
problem Evaluating the impact of flipped classrooms on higher education outcomes.
method Double/debiased machine learning (DML) approach to analyze student data.
result No significant positive effects on exam scores, passing rates, or knowledge retention.
Batch Normalization is a commonly used trick to improve the training of deep neural networks. These neural networks use L2 regularization, also called weight decay, ostensibly to prevent overfitting. However, we show that L2 regularization has no regularizing effect when combined with normalization. Instead, regulariza…
Adaptive learning rates improve FTPL's BOBW guarantees in bandit problems.
problem Improving Follow-the-Perturbed-Leader's BOBW guarantees in bandit problems.
method Introducing surrogate probability functions to compute adaptive learning rates without exact probabilities.
result BOBW guarantees for FTPL with Pareto perturbations for any α>1. E-Commerce (E-Com) search is an emerging important new application of information retrieval. Learning to Rank (LETOR) is a general effective strategy for optimizing search engines, and is thus also a key technology for E-Com search. While the use of LETOR for web search has been well studied, its use for E-Com search h…
Study reveals dynamics of neural networks with normalization, weight decay, and SGD.
problem Understanding the equilibrium condition in Spherical Motion Dynamics (SMD).
method Investigates SMD by exploring the cause of equilibrium condition, introducing assumptions, proposing angular update, and verifying theoretical results.
result Proves weight norm and angular update can converge at linear rate under given assumptions.
The paper presents a multi-power law for predicting loss curves across different learning rate schedules.
problem Understanding and optimizing the relationship between model performance and hyperparameters, especially learning rates.
method Proposes a multi-power law that combines power laws based on the sum of learning rates and additional laws for loss reduction due to decay.
result The multi-power law accurately predicts loss curves for unseen learning rate schedules and finds a schedule that outperforms cosine learning rate.
Proposes a learning rate method for shallow nets based on gradient Lipschitz constant.
problem Finding optimal learning rates for shallow neural networks.
method Associates learning rate with gradient Lipschitz constant and proposes a search algorithm.
result The proposed method significantly outperforms existing tuning methods.
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.
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…
Stochastic Gradient Descent shows directional bias with moderate learning rates, impacting optimization outcomes.
problem Understanding the bias of SGD with moderate learning rates in practical scenarios.
method Analyzing SGD and GD on an overparameterized linear regression problem.
result SGD converges along large eigenvalue directions, GD along small ones, affecting early stopping outcomes.
Deep neural networks perform well due to optimization effects.
problem Understanding why deep neural networks perform well empirically.
method Proved convergence theorem for nonconvex optimization, applied to learning problems.
result Wide deep neural networks converge to global optima with gradient descent.