We study a budgeted hyper-parameter tuning problem, where we optimize the tuning result under a hard resource constraint. We propose to solve it as a sequential decision making problem, such that we can use the partial training progress of configurations to dynamically allocate the remaining budget. Our algorithm combi…
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.
168,742 papers · 148 categories
Trend · papers per month
48 results for “Budget Tuning”
SIREN protocol corrects optimistic winner's scores in LLM evaluation.
problem Optimistic winner's scores in LLM evaluation due to adaptive benchmarking.
method SIREN protocol that freezes post-search shortlist, separates selection and evaluation, and uses bootstrap for uncertainty quantification.
result SIREN provides valid confidence intervals for procedure-performance curves and deployment conclusions.
Finding the Sweet Spot: Trading Quality, Cost, and Speed During Inference-Time LLM Reflectionstat.ML
This paper compares self-reflection and budget tuning for LLMs, revealing domain-specific performance gains.
problem Improving inference-time performance of LLMs without retraining, balancing quality, cost, and latency.
method Systematic comparison of self-reflection and budget tuning across mathematical reasoning and translation tasks, evaluating various LLMs and model families.
result Substantial domain-dependent variation in self-reflection effectiveness, with up to 220% performance gains in mathematical reasoning.
AutoML explores vs. exploits promising classifiers to improve performance.
problem Maximizing ML pipeline performance within limited time and resource constraints.
method Empirical study comparing exploiting vs. exploring the search space for promising classifiers.
result Exploiting the most promising classifiers does not statistically improve pipeline performance.
HAMLET optimizes algorithm selection for machine learning tasks.
problem Limited time budgets and computational resources make traditional bandit approaches ineffective for automated algorithm selection.
method HAMLET incorporates learning curve extrapolation and time-awareness to select machine learning algorithms.
result HAMLET variants outperform other bandit-based strategies in experiments with recorded hyperparameter tuning traces.
Algorithm allocates budgets to tasks with semi-bandit feedback, achieving near-optimal regret bounds.
problem Stochastic budget allocation with censored semi-bandit feedback.
method Optimism-based algorithm operating under censored semi-bandit feedback.
result Regret scales polylogarithmically with horizon T in diminishing-returns regimes.
Framework ranks sectors influenced by Indian Union Budgets.
problem Real-time analysis of budgetary impacts on sector-specific equity performance.
method Fine-tuned embeddings and language models for sector identification and performance ranking.
result 0.997 NDCG score in predicting sector ranks based on post-budget performances.
StatLoRA uses statistical inference to allocate ranks in LoRA fine-tuning, improving performance.
problem Balancing efficiency, expressiveness, and generalization in LoRA rank allocation.
method Formulates LoRA rank allocation as a statistical hypothesis testing problem, using estimated p-values to determine component retention or pruning.
result StatLoRA achieves comparable or better performance than existing methods under matched rank budgets.
New algorithms improve privacy and utility of large language models.
problem Privacy-preserving fine-tuning of large language models.
method Meta-framework for differentially private fine-tuning, inspired by recent success in fine-tuning.
result Private fine-tuned models achieve utility close to non-private models, with improved privacy and efficiency.
We consider the problem of minimizing a convex risk with stochastic subgradients guaranteeing -locally differentially private (-LDP). While it has been shown that stochastic optimization is possible with -LDP via the standard SGD (Song et al., 2013), its convergence rate largely depends on the learning rate, w…
FisherSFT selects informative examples to fine-tune LLMs efficiently.
problem Adapting large language models to new domains efficiently.
method Selects examples maximizing information gain using Hessian of log-likelihood.
result Empirically demonstrates improved performance with reduced computational cost.
New method FedEx accelerates federated hyperparameter tuning.
problem Federated hyperparameter tuning challenges in distributed learning.
method FedEx method connecting to weight-sharing, adapted for federated optimization.
result FedEx outperforms natural baselines on various benchmarks.
New method optimizes costly functions with unknown costs and budget constraints.
problem Optimizing functions with unknown and heterogeneous evaluation costs under a budget constraint.
method Budgeted multi-step expected improvement acquisition function.
result Our method outperforms existing approaches in various synthetic and real problems.
Paper simplifies DP composition for adaptive privacy budgets, enabling better privacy and accuracy in deep learning.
problem Tension between efficiency and flexibility in DP composition theorems.
method Rényi Differential Privacy (RDP) for adaptive privacy budgets, proving simpler composition theorem with smaller constants.
result Practical DP composition for adaptive privacy budgets, enabling better privacy and accuracy in deep learning.
The performance of optimizers, particularly in deep learning, depends considerably on their chosen hyperparameter configuration. The efficacy of optimizers is often studied under near-optimal problem-specific hyperparameters, and finding these settings may be prohibitively costly for practitioners. In this work, we arg…
Automated hyperparameter tuning aspires to facilitate the application of machine learning for non-experts. In the literature, different optimization approaches are applied for that purpose. This paper investigates the performance of Differential Evolution for tuning hyperparameters of supervised learning algorithms for…
Paper introduces a privacy-preserving line search method for optimization.
problem Optimization performance depends on step size tuning, which is difficult and privacy-sensitive.
method Introduces a stochastic adaptive line search algorithm that satisfies differential privacy.
result The algorithm efficiently uses privacy budget and outperforms existing private optimizers.
Paper proposes EEIPU, a memoization-aware BO algorithm to reduce hyperparameter tuning costs.
problem High costs in GPU-days for training and fine-tuning language models.
method Memoization-aware Bayesian Optimization (EEIPU) algorithm in tandem with pipeline caching.
result EEIPU produces 103% more hyperparameter candidates and 108% more validation metric improvement.
The paper tackles the issue of preferential attachment in targeted display advertising by developing domain-adaptation approaches.
problem Skewed distribution of data leads to preferential attachment towards high-budget partners.
method Develops domain-adaptation approaches to predict interested users for low-budget partners.
result Proposed approaches outperform other domain-adaptation methods across different points of campaigns.
EB-TCε identifies the best arm with ε confidence in stochastic bandits.
problem Identifying the best arm in stochastic bandits with a fixed level of confidence.
method EB-TCε is a novel sampling rule for ε-best arm identification in stochastic bandits.
result EB-TCε is the first anytime algorithm for fixed confidence or fixed budget identification.
Reinforcement Learning for Non-Stationary Markov Decision Processes: The Blessing of (More) Optimismcs.LG
New RL method tackles dynamic MDPs with evolving rewards and states.
problem Dynamic MDPs with evolving rewards and states.
method Sliding Window Upper-Confidence bound for Reinforcement Learning (SWUCRL2-CW) and Bandit-over-Reinforcement Learning (BORL).
result Achieves dynamic regret bound for non-stationary MDPs.
GeLoRA optimizes LoRA fine-tuning by dynamically adjusting ranks based on intrinsic dimensionality.
problem Efficient fine-tuning of large language models with limited computational resources.
method GeLoRA computes intrinsic dimensionality to adaptively select LoRA ranks, balancing expressivity and efficiency.
result GeLoRA consistently outperforms recent baselines within the same parameter budget on multiple tasks.
EVA adapts LoRA for faster, more efficient fine-tuning.
problem Fast and efficient fine-tuning of large models for specific tasks.
method EVA uses directions capturing most activation variance for initialization, maximizing gradient signal and reducing parameters.
result EVA achieves faster convergence and higher average scores across tasks, reducing parameters.
LOFT separates subspace rotation and transformation for orthogonal fine-tuning.
problem Conflating subspace rotation and transformation in orthogonal fine-tuning.
method LOFT explicitly separates subspace rotation and transformation, using task-aware support selection.
result LOFT recovers principal-subspace orthogonal adaptation and improves efficiency-performance trade-off.
New optimizer improves privacy-protected hyperparameter tuning.
problem No practical methods for differentially private hyperparameter selection.
method Study honest hyperparameter selection under DP, show adaptive optimizers like DPAdam have an advantage.
result DPAdam optimizes hyperparameters more efficiently under DP constraints.
FGSM is more stable in adversarially robust transfer learning than PGD.
problem Computational efficiency in adversarially robust transfer learning.
method Revisited use of FGSM in adversarial fine-tuning.
result FGSM is more stable and efficient in adversarial fine-tuning.
PDO optimizes LLM prompts without labels, improving performance.
problem Optimizing prompts for LLMs without access to labeled data.
method Pairwise preference feedback, dueling bandits, Thompson Sampling, mutation.
result PDO identifies stronger prompts than label-free methods.
Adversarial training shows promise as an approach for training models that are robust towards adversarial perturbation. In this paper, we explore some of the practical challenges of adversarial training. We present a sensitivity analysis that illustrates that the effectiveness of adversarial training hinges on the sett…
Improves privacy guarantees by analyzing randomness in privacy-preserving mechanisms.
problem Balancing user privacy and business constraints in privacy-preserving mechanisms.
method Analyzes explicit and implicit randomness in privacy mechanisms and proposes a probabilistic calibration method.
result Proposes privacy at risk, providing stronger privacy guarantees with quantifiable risks.
New method upscales models and transfers hyperparameters efficiently.
problem Efficiently scaling and tuning large neural networks.
method Introduces a general upscaling method and extends Transfer for hyperparameter tuning.
result Effective hyperparameter transfer for upscaled models.
We introduce algorithms that achieve state-of-the-art \emph{dynamic regret} bounds for non-stationary linear stochastic bandit setting. It captures natural applications such as dynamic pricing and ads allocation in a changing environment. We show how the difficulty posed by the non-stationarity can be overcome by a nov…
Tabular Data Generation Models: An In-Depth Survey and Performance Benchmarks with Extensive Tuningcs.LG
This study benchmarks tabular data generation models, optimizing hyperparameters and feature encodings.
problem Generating realistic tabular data is challenging due to heterogeneity, non-smooth distributions, and complex dependencies.
method Comprehensive evaluation of five model families on 16 datasets, considering hyperparameters, feature encodings, and architectures.
result Large-scale dataset-specific tuning significantly improves model performance, especially for diffusion-based models.
Automated Budget Constrained Training optimizes model training under time constraints.
problem Balancing model quality and computational cost in constrained time.
method Developed a hyperparameter optimisation algorithm that learns the relationship between hyperparameters, model quality, and computational cost.
result The algorithm optimally decides whether to terminate or continue training, and what hyperparameters to use.
Differentially private learning on real-world data poses challenges for standard machine learning practice: privacy guarantees are difficult to interpret, hyperparameter tuning on private data reduces the privacy budget, and ad-hoc privacy attacks are often required to test model privacy. We introduce three tools to ma…
This work refines grid size selection for non-interactive private -means clustering.
problem Choosing the optimal number of grids for privatized -means clustering.
method Proposes a refined grid-size selection rule to minimize expected deviation in the K-means objective function.
result The proposed strategy results in more accurate clustering compared to prior work, even under tight privacy budgets.
GNMR controls runtime stability in low-precision language model training.
problem Efficient low-precision training faces numerical risks at specific operators.
method GNMR compares gradient norms to historical means, applying bounded recovery actions.
result GNMR preserves high-fidelity quality with sparse, budgeted recovery.
This paper considers a multi-armed bandit game where the number of arms is much larger than the maximum budget and is effectively infinite. We characterize necessary and sufficient conditions on the total budget for an algorithm to return an ε-good arm with probability at least 1 - δ. In such situations, the sample com…
Quantum kernels show no advantage in stock return prediction, but differ in stability metrics.
problem Determining if quantum kernels improve stock return prediction.
method Controlled horse race on Chinese A-share market with identical training subsamples and tuning budgets.
result Quantum kernels do not outperform classical RBF controls in cross-sectional stock return prediction.
Efficient and Unbiased Sampling from Boltzmann Distributions via Variance-Tuned Diffusion Modelscs.LG
VT-DIS improves sampling from Boltzmann distributions with minimal overhead.
problem Bias in Monte Carlo estimates from score-based diffusion models.
method Variance-Tuned Diffusion Importance Sampling (VT-DIS) adapts noise covariance to correct bias.
result VT-DIS achieves effective sample sizes of 80%, 35%, and 3.5% on benchmarks, using less computational budget.
High sensitivity of neural architecture search (NAS) methods against their input such as step-size (i.e., learning rate) and search space prevents practitioners from applying them out-of-the-box to their own problems, albeit its purpose is to automate a part of tuning process. Aiming at a fast, robust, and widely-appli…
POET enables large neural network training on tiny devices with reduced energy.
problem Training large neural networks on memory-limited edge devices.
method Jointly optimizes rematerialization and paging for memory reduction, formulating an MILP for energy-efficient training.
result POET trains ResNet-18 and BERT within Cortex-M memory constraints, outperforming current methods in energy efficiency.
Noise in SGD helps deep nets generalize better, even with smaller batch sizes.
problem The generalization benefit of using noise in SGD over large batch sizes.
method Carefully designed experiments and rigorous hyperparameter sweeps on various models.
result Small or moderately large batch sizes outperform very large batches on test sets.
Improved sampling efficiency for molecular systems using path gradients after Flow Matching.
problem Improving sampling efficiency for complex molecular systems.
method Hybrid approach combining Flow Matching and path gradients.
result Up to a threefold increase in sampling efficiency for molecular systems.
POCAII optimizes hyperparameters with a new approach, showing superior performance.
problem Hyperparameter optimization with limited resources.
method Explicitly separates search and evaluation phases, focusing on exploration and exploitation.
result POCAII outperforms state-of-the-art HPO algorithms in low-budget scenarios.
New study finds optimal hyperparameter tuning crucial for fair optimizer comparisons.
problem Inadequate hyperparameter tuning and misleading evaluation setups hinder fair comparisons of optimizers.
method Systematic study of ten optimizers across four model scales and data-to-model ratios.
result Optimal hyperparameters for one optimizer may be suboptimal for another, and many claimed speedups are lower than expected.
The paper develops a theory for iterative self-improvement of models, proving conditions for better performance with easy-to-hard curricula.
problem Lack of theoretical foundation for iterative self-improvement in practical settings.
method Modeling self-improvement as maximum-likelihood fine-tuning on reward-filtered distributions and proving finite-sample guarantees.
result Explicit feedback loop and conditions for better performance with easy-to-hard curricula.
Machine learning (ML) problems are often posed as highly nonlinear and nonconvex unconstrained optimization problems. Methods for solving ML problems based on stochastic gradient descent are easily scaled for very large problems but may involve fine-tuning many hyper-parameters. Quasi-Newton approaches based on the lim…
We consider un-discounted reinforcement learning (RL) in Markov decision processes (MDPs) under temporal drifts, ie, both the reward and state transition distributions are allowed to evolve over time, as long as their respective total variations, quantified by suitable metrics, do not exceed certain variation budgets. …