Proposes glocal hypergradient estimation for hyperparameter optimization.
problem Combining reliability and efficiency in hyperparameter optimization.
method Uses Koopman operator theory to approximate global hypergradients from local ones.
result Achieves both reliability and efficiency in hyperparameter optimization.
One-pass optimisation for high-dimensional hyperparameters.
problem Efficient optimisation of hyperparameters in machine learning models.
method Approximate hypergradient-based optimisation for any continuous hyperparameter, requiring only one training episode.
result Competitive performance on various datasets without hyperparameter restarts.
Efficiently optimizes hyperparameters for PDE and inverse problems using Gaussian processes.
problem Hyperparameter optimization for scientific computing and inference methods.
method Bilevel optimization with Gauss-Newton linearization for efficient hyperparameter updates.
result Significant improvements in accuracy and robustness compared to random initialization.
PES method reduces bias in gradient estimation for unrolled graphs.
problem High variance and bias in gradient estimation for unrolled computation graphs.
method Divide graph into unrolls, apply ES update, accumulate correction terms.
result PES provides unbiased, low-variance gradient estimates.
This paper quantifies hyperparameter transfer and finds embedding layer learning rate is key.
problem Quantifying optimal hyperparameters for large language models across scales.
method Developed three metrics to quantify hyperparameter transfer and investigated the importance of embedding layer learning rate.
result Maximal Update (μP) parameterization offers high-quality learning rate transfer compared to standard parameterization (SP).
Sparse pseudo-point approximations for Gaussian process (GP) models provide a suite of methods that support deployment of GPs in the large data regime and enable analytic intractabilities to be sidestepped. However, the field lacks a principled method to handle streaming data in which both the posterior distribution ov…
A new principle for optimizer selection improves training speed and performance.
problem Finding the best optimizer hyperparameters for faster training.
method Formulate optimizer selection as maximizing the expected drop rate in loss, treating gradients and updates as signals and an optimizer as a causal filter.
result Greedy optimizer selection yields stable and effective momentum rules.
This work analyzes how often to update the target network in Q-learning.
problem Understanding the optimal frequency of target network updates in Q-learning.
method Formulated target updates as a nested optimization scheme, derived finite-time convergence analysis.
result Optimal target update frequency increases geometrically over time.
OEUVRE estimates online loss with constant time and memory, outperforming other methods.
problem Accurately estimating expected loss in online learning.
method Recursive evaluation of each sample on current and previous models, using algorithmic stability for updates.
result Consistency, convergence rates, and concentration bounds proved for OEUVRE.
The study analyzes deep linear networks from random initialization, capturing dynamics and hyperparameter effects.
problem Understanding training dynamics in deep linear networks from random initialization.
method Theoretical analysis of gradient descent dynamics in deep linear networks with random initialization and large data.
result Captures the 'wider is better' effect and hyperparameter transfer effects, contrasting with neural-tangent parameterization.
Hyperparameter optimization can be formulated as a bilevel optimization problem, where the optimal parameters on the training set depend on the hyperparameters. We aim to adapt regularization hyperparameters for neural networks by fitting compact approximations to the best-response function, which maps hyperparameters …
Most models in machine learning contain at least one hyperparameter to control for model complexity. Choosing an appropriate set of hyperparameters is both crucial in terms of model accuracy and computationally challenging. In this work we propose an algorithm for the optimization of continuous hyperparameters using in…
The performance of policy gradient methods is sensitive to hyperparameter settings that must be tuned for any new application. Widely used grid search methods for tuning hyperparameters are sample inefficient and computationally expensive. More advanced methods like Population Based Training that learn optimal schedule…
Optimization algorithms with momentum, e.g., (ADAM), have been widely used for building deep learning models due to the faster convergence rates compared with stochastic gradient descent (SGD). Momentum helps accelerate SGD in the relevant directions in parameter updating, which can minify the oscillations of parameter…
New method optimizes hyperparameters for non-smooth problems efficiently.
problem Efficiently tuning hyperparameters for non-smooth cost functions.
method Combines hyperparameter search with proximal gradient updates.
result Method converges to local optimum of LOO validation error.
Scalable tools for nested optimization in deep learning.
problem Solving nested optimization problems on a large scale in deep learning.
method Building scalable tools for bilevel optimization.
result Tools for nested optimization scale to deep learning setups.
Conditional kernel mean embeddings are nonparametric models that encode conditional expectations in a reproducing kernel Hilbert space. While they provide a flexible and powerful framework for probabilistic inference, their performance is highly dependent on the choice of kernel and regularization hyperparameters. Neve…
EGAB algorithms improve online portfolio selection.
problem Online portfolio selection problem.
method Generalized exponentiated gradient (EG) updates with Alpha-Beta divergence regularization.
result EGAB algorithms enhance portfolio performance, especially with transaction costs.
We study two procedures (reverse-mode and forward-mode) for computing the gradient of the validation error with respect to the hyperparameters of any iterative learning algorithm such as stochastic gradient descent. These procedures mirror two methods of computing gradients for recurrent neural networks and have differ…
System learns optimizer hyperparameters to generalize across tasks.
problem Lack of generalization in learning optimizers for neural networks.
method Generalization-first approach, learning optimizer hyperparameters.
result System outperforms Adam on various tasks, including unseen modalities.
Aims to describe neural network training dynamics using two-time-scale models.
problem Lack of a general mathematical description of neural network training.
method Introduces a theoretical framework based on two-time-scale population dynamics.
result Derives selection-mutation equations and effective fitness for hyperparameters.
New framework explains fast transfer of hyperparameters across model scales.
problem Understanding and optimizing hyperparameters for large-scale models.
method Developed a conceptual framework for HP transfer across scale, showing fast transfer is equivalent to useful transfer for compute-optimal grid search.
result Fast transfer of hyperparameters is equivalent to useful transfer for compute-optimal grid search, offering asymptotic computational advantage.
Efficient algorithm for mobile health provides timely physical activity suggestions.
problem Inefficient reinforcement learning for mobile health settings.
method Contextual bandit algorithm with linear mixed effects model and hyper-parameter updating.
result Improves speed and accuracy by up to 99% and 56%.
Derives a family of hyperparameter scaling strategies for neural networks.
problem Optimizing hyperparameters for wide and deep neural networks.
method Introduces a one-parameter family of hyperparameter scaling strategies.
result Reveals proper scaling of depth with width for large-scale models.
LGD algorithm learns optimal hyperparameters for regression tasks.
problem Learning to learn hyperparameters for regression problems.
method Langevin Gradient Descent (LGD) approximates posterior distribution.
result Meta-learning optimal hyperparameters achieves Bayes' optimal solution.
New MD algorithms using Tempesta logarithms for machine learning.
problem Optimization in machine learning with tailored hyperparameters.
method Developed Mirror Descent algorithms using Tempesta multi-parametric logarithms.
result Wide and flexible family of Mirror Descent and mirror-less updates.
New method improves uncertainty quantification in latent variable models.
problem Uncertainty quantification in latent variable models with SGLD-Gibbs.
method Statistical scaling limit theory for SGLD-Gibbs, proposing hyperparameter tuning.
result Explicit guidance on hyperparameter tuning for SGLD-Gibbs ensures meaningful uncertainty quantification.
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.
New method μP2 improves neural network training by scaling perturbations layerwise.
problem Improving neural network performance as models scale up.
method Layerwise perturbation scaling in the infinite-width limit of neural networks.
result Layerwise perturbation scaling ensures all layers are effectively perturbed in the limit.
Many of the recent triumphs in machine learning are dependent on well-tuned hyperparameters. This is particularly prominent in reinforcement learning (RL) where a small change in the configuration can lead to failure. Despite the importance of tuning hyperparameters, it remains expensive and is often done in a naive an…
Stochastic variational inference allows for fast posterior inference in complex Bayesian models. However, the algorithm is prone to local optima which can make the quality of the posterior approximation sensitive to the choice of hyperparameters and initialization. We address this problem by replacing the natural gradi…
New method dynamically adjusts UTD ratio to balance under- and overfitting in RL.
problem Balancing under- and overfitting in world model learning for RL.
method Dynamic adjustment of UTD ratio based on validation performance on a small subset of experience data.
result Our method improves balance between under- and overfitting compared to default settings and competitive with extensive hyperparameter search.
Automates fairness and accuracy optimization in deep learning models for tabular data.
problem Improving fairness and accuracy in neural models for tabular data.
method Employed multi-objective Neural Architecture Search (NAS) and Hyperparameter Optimization (HPO) to find new models.
result Jointly optimized architectures that consistently outperform single-objective fairness mitigation methods.
Bayesian sparse learning method improves deep neural network efficiency.
problem Sparse learning in deep neural networks with complex geometry.
method Preconditioned stochastic gradient Langevin Dynamics (PSGLD) for sampling and adaptive optimization of hyperparameters.
result The proposed algorithm achieves asymptotic convergence with controlled bias.
New recommendations improve Gaussian process accuracy and stability.
problem Numerical instabilities and poor test likelihoods in iterative Gaussian process learning.
method Investigated CG tolerance, preconditioner rank, and Lanczos decomposition rank. Recommended small CG tolerance and large root decomposition size.
result L-BFGS-B optimizer achieves convergence with fewer gradient updates, improving Gaussian process accuracy.
Test-time training adapts a pretrained model to each prompt via parameter updates, improving accuracy under pretraining-to-test distribution shifts.
problem Improving accuracy of pretrained models under distribution shifts.
method Explaining TTT behavior through a decision-theoretic lens.
result TTT reduces prediction error when updates are spectrally matched to the prompt's signal-to-noise ratio and aligned with query-relevant eigen-directions.
FL+HC improves federated learning on non-iid data by clustering local updates.
problem FL struggles with non-iid data, leading to suboptimal models.
method Introduce hierarchical clustering to separate and train clusters of clients independently.
result FL+HC converges faster and achieves higher accuracy than standard FL.
This study compares 6 imitation learning algorithms using a common dataset and hyperparameter budget.
problem Difficulty in comparing different imitation learning algorithms due to varying datasets, base RL algorithms, and evaluation settings.
method Reimplemented and updated 6 different imitation learning algorithms, using a common off-policy algorithm (SAC) and a widely-used dataset (D4RL). Evaluated on a range of expert trajectories.
result GAIL consistently performs well across different sample sizes, while AdRIL performs well with one important hyperparameter to tune and behavioral cloning remains a strong baseline when data is plentiful.
TCP provides well-calibrated prediction intervals for nonstationary time series.
problem Nonstationary time series forecasting with well-calibrated prediction intervals.
method Temporal Conformal Prediction (TCP) couples a modern quantile forecaster with a rolling split-conformal calibration layer.
result TCP achieves near-nominal coverage, providing slightly wider intervals than Historical Simulation.
We consider online forecasting problems for non-convex machine learning models. Forecasting introduces several challenges such as (i) frequent updates are necessary to deal with concept drift issues since the dynamics of the environment change over time, and (ii) the state of the art models are non-convex models. We ad…
Traditionally, deep learning algorithms update the network weights whereas the network architecture is chosen manually, using a process of trial and error. In this work, we propose two novel approaches that automatically update the network structure while also learning its weights. The novelty of our approach lies in o…
Iterative procedures for parameter estimation based on stochastic gradient descent allow the estimation to scale to massive data sets. However, in both theory and practice, they suffer from numerical instability. Moreover, they are statistically inefficient as estimators of the true parameter value. To address these tw…
Statistical analysis of regularization in continual learning tasks.
problem Understanding how regularization affects model performance in sequential learning.
method Derivation of convergence rates, iterative update formula, and optimal hyperparameters for generalized ℓ2-regularization.
result Optimal hyperparameters balance forward and backward knowledge transfer, improving model performance.
We aim to design adaptive online learning algorithms that take advantage of any special structure that might be present in the learning task at hand, with as little manual tuning by the user as possible. A fundamental obstacle that comes up in the design of such adaptive algorithms is to calibrate a so-called step-size…
Existing approaches for training neural networks with user-level differential privacy (e.g., DP Federated Averaging) in federated learning (FL) settings involve bounding the contribution of each user's model update by clipping it to some constant value. However there is no good a priori setting of the clipping norm acr…
We propose a new method for blind system identification. Resorting to a Gaussian regression framework, we model the impulse response of the unknown linear system as a realization of a Gaussian process. The structure of the covariance matrix (or kernel) of such a process is given by the stable spline kernel, which has b…
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.
The ever increasing computational cost of Deep Neural Networks (DNN) and the demand for energy efficient hardware for DNN acceleration has made accuracy and hardware cost co-optimization for DNNs tremendously important, especially for edge devices. Owing to the large parameter space and cost of evaluating each paramete…