Scaling laws are powerful tools to predict the performance of large language models. However, current scaling laws fall short of accounting for inference costs. In this work, we first show that model architecture affects inference latency, where models of the same size can have up to 3.5x difference in latency. To tackle this challenge, we modify the Chinchilla scaling laws to co-optimize the model parameter count, the number of training tokens, and the model architecture. Due to the reason that models of similar training loss exhibit gaps in downstream evaluation, we also propose a novel method to train inference-efficient models based on the revised scaling laws. We perform extensive empirical studies to fit and evaluate our inference-aware scaling laws. We vary model parameters from 80M to 1B, training tokens from 1.6B to 30B, and model shapes, training a total of 63 models. Guided by our inference-efficient scaling law and model selection method, we release the Morph-1B model, which improves inference latency by 1.8x while maintaining accuracy on downstream tasks compared to open-source models, pushing the Pareto frontier of accuracy-latency tradeoff.
Modern deep neural networks display striking examples of rich internal computational structure. Uncovering principles governing the development of such structure is a priority for the science of deep learning. In this paper, we explore the transient ridge phenomenon: when transformers are trained on in-context linear regression tasks with intermediate task diversity, they initially behave like ridge regression before specializing to the tasks in their training distribution. This transition from a general solution to a specialized solution is revealed by joint trajectory principal component analysis. Further, we draw on the theory of Bayesian internal model selection to suggest a general explanation for the phenomena of transient structure in transformers, based on an evolving tradeoff between loss and complexity. We empirically validate this explanation by measuring the model complexity of our transformers as defined by the local learning coefficient.
Choosing models from a well-fitted evolved population that generalizes beyond training data is difficult. We introduce a pragmatic method to estimate model complexity using Hessian rank for post-processing selection. Complexity is approximated by averaging the model output Hessian rank across a few points (N=3), offering efficient and accurate rank estimates. This method aligns model selection with input data complexity, calculated using intrinsic dimensionality (ID) estimators. Using the StackGP system, we develop symbolic regression models for the Penn Machine Learning Benchmark and employ twelve scikit-dimension library methods to estimate ID, aligning model expressiveness with dataset ID. Our data-informed complexity metric finds the ideal complexity window, balancing model expressiveness and accuracy, enhancing generalizability without bias common in methods reliant on user-defined parameters, such as parsimony pressure in weight selection.
Large Language Models (LLMs) have shown remarkable capabilities across various natural language processing tasks but often struggle to excel uniformly in diverse or complex domains. We propose a novel ensemble method - Diverse Fingerprint Ensemble (DFPE), which leverages the complementary strengths of multiple LLMs to achieve more robust performance. Our approach involves: (1) clustering models based on response "fingerprints" patterns, (2) applying a quantile-based filtering mechanism to remove underperforming models at a per-subject level, and (3) assigning adaptive weights to remaining models based on their subject-wise validation accuracy. In experiments on the Massive Multitask Language Understanding (MMLU) benchmark, DFPE outperforms the best single model by 3% overall accuracy and 5% in discipline-level accuracy. This method increases the robustness and generalization of LLMs and underscores how model selection, diversity preservation, and performance-driven weighting can effectively address challenging, multi-faceted language understanding tasks.
Machine learning models are widely used in applications where reliability and robustness are critical. Model evaluation often relies on single-point estimates of performance metrics such as accuracy, F1 score, or mean squared error, that fail to capture the inherent variability in model performance. This variability arises from multiple sources, including train-test split, weights initialization, and hyperparameter tuning. Investigating the characteristics of performance metric distributions, rather than focusing on a single point only, is essential for informed decision-making during model selection and optimization, especially in high-stakes settings. How does the performance metric vary due to intrinsic uncertainty in the selected modeling approach? For example, train-test split is modified, initial weights for optimization are modified or hyperparameter tuning is done using an algorithm with probabilistic nature? This is shifting the focus from identifying a single best model to understanding a distribution of the performance metric that captures variability across different training conditions. By running multiple experiments with varied settings, empirical distributions of performance metrics can be generated. Analyzing these distributions can lead to more robust models that generalize well across diverse scenarios. This contribution explores the use of quantiles and confidence intervals to analyze such distributions, providing a more complete understanding of model performance and its uncertainty. Aimed at a statistically interested audience within the machine learning community, the suggested approaches are easy to implement and apply to various performance metrics for classification and regression problems. Given the often long training times in ML, particular attention is given to small sample sizes (in the order of 10-25).
Foundation models are widely employed in medical image analysis, due to their high adaptability and generalizability for downstream tasks. With the increasing number of foundation models being released, model selection has become an important issue. In this work, we study the capabilities of foundation models in medical image classification tasks by conducting a benchmark study on the MedMNIST dataset. Specifically, we adopt various foundation models ranging from convolutional to Transformer-based models and implement both end-to-end training and linear probing for all classification tasks. The results demonstrate the significant potential of these pre-trained models when transferred for medical image classification. We further conduct experiments with different image sizes and various sizes of training data. By analyzing all the results, we provide preliminary, yet useful insights and conclusions on this topic.
The transmission dynamics of an epidemic are rarely homogeneous. Super-spreading events and super-spreading individuals are two types of heterogeneous transmissibility. Inference of super-spreading is commonly carried out on secondary case data, the expected distribution of which is known as the offspring distribution. However, this data is seldom available. Here we introduce a multi-model framework fit to incidence time-series, data that is much more readily available. The framework consists of five discrete-time, stochastic, branching-process models of epidemics spread through a susceptible population. The framework includes a baseline model of homogeneous transmission, a unimodal and a bimodal model for super-spreading events, as well as a unimodal and a bimodal model for super-spreading individuals. Bayesian statistics is used to infer model parameters using Markov Chain Monte-Carlo. Model comparison is conducted by computing Bayes factors, with importance sampling used to estimate the marginal likelihood of each model. This estimator is selected for its consistency and lower variance compared to alternatives. Application to simulated data from each model identifies the correct model for the majority of simulations and accurately infers the true parameters, such as the basic reproduction number. We also apply our methods to incidence data from the 2003 SARS outbreak and the Covid-19 pandemic. Model selection consistently identifies the same model and mechanism for a given disease, even when using different time series. Our estimates are consistent with previous studies based on secondary case data. Quantifying the contribution of super-spreading to disease transmission has important implications for infectious disease management and control. Our modelling framework is disease-agnostic and implemented as an R package, with potential to be a valuable tool for public health.
Time series analysis has become crucial in various fields, from engineering and finance to healthcare and social sciences. In this paper, we present a comprehensive review and evaluation of time series embedding methods for effective representations in machine learning and deep learning models. We introduce a taxonomy of embedding techniques, categorizing them based on their theoretical foundations and application contexts. Unlike previous surveys, our work provides a quantitative evaluation of representative methods from each category by assessing their performance on downstream classification tasks across diverse real-world datasets. Our experimental results demonstrate that the performance of embedding methods varies significantly depending on the dataset and classification algorithm used, highlighting the importance of careful model selection and extensive experimentation for specific applications, including engineering systems. To facilitate further research and practical applications, we provide an open-source code repository implementing these embedding methods. This study contributes to the field by offering a systematic comparison of time series embedding techniques, guiding practitioners in selecting appropriate methods for their specific applications, and providing a foundation for future advancements in time series analysis.
We propose a unified framework for automatic debiased machine learning (autoDML) to perform inference on smooth functionals of infinite-dimensional M-estimands, defined as population risk minimizers over Hilbert spaces. By automating debiased estimation and inference procedures in causal inference and semiparametric statistics, our framework enables practitioners to construct valid estimators for complex parameters without requiring specialized expertise. The framework supports Neyman-orthogonal loss functions with unknown nuisance parameters requiring data-driven estimation, as well as vector-valued M-estimands involving simultaneous loss minimization across multiple Hilbert space models. We formalize the class of parameters efficiently estimable by autoDML as a novel class of nonparametric projection parameters, defined via orthogonal minimum loss objectives. We introduce three autoDML estimators based on one-step estimation, targeted minimum loss-based estimation, and the method of sieves. For data-driven model selection, we derive a novel decomposition of model approximation error for smooth functionals of M-estimands and propose adaptive debiased machine learning estimators that are superefficient and adaptive to the functional form of the M-estimand. Finally, we illustrate the flexibility of our framework by constructing autoDML estimators for the long-term survival under a beta-geometric model.
Complex dynamical systems, from macromolecules to ecosystems, are often modeled by stochastic differential equations. To learn such models from data, a common approach involves sparse selection among a large function library. However, we show that overfitting arises - not just from individual model complexity, but also from the combinatorial growth of possible models. To address this, we introduce Parsimonious Stochastic Inference (PASTIS), a principled method combining likelihood-estimation statistics with extreme value theory to suppress superfluous parameters. PASTIS outperforms existing methods and reliably identifies minimal models, even with low sampling rates or measurement error. It extends to stochastic partial differential equations, and applies to ecological networks and reaction-diffusion dynamics.
In high-dimensional regression, feature selection methods, such as sequential feature selection (SeqFS), are commonly used to identify relevant features. When data is limited, domain adaptation (DA) becomes crucial for transferring knowledge from a related source domain to a target domain, improving generalization performance. Although SeqFS after DA is an important task in machine learning, none of the existing methods can guarantee the reliability of its results. In this paper, we propose a novel method for testing the features selected by SeqFS-DA. The main advantage of the proposed method is its capability to control the false positive rate (FPR) below a significance level $\alpha$ (e.g., 0.05). Additionally, a strategic approach is introduced to enhance the statistical power of the test. Furthermore, we provide extensions of the proposed method to SeqFS with model selection criteria including AIC, BIC, and adjusted R-squared. Extensive experiments are conducted on both synthetic and real-world datasets to validate the theoretical results and demonstrate the proposed method's superior performance.
Repeated waves of emerging variants during the SARS-CoV-2 pandemics have highlighted the urge of collecting longitudinal genomic data and developing statistical methods based on time series analyses for detecting new threatening lineages and estimating their fitness early in time. Most models study the evolution of the prevalence of particular lineages over time and require a prior classification of sequences into lineages. Such process is prone to induce delays and bias. More recently, few authors studied the evolution of the prevalence of mutations over time with alternative clustering approaches, avoiding specific lineage classification. Most of the aforementioned methods are however either non parametric or unsuited to pooled data characterizing, for instance, wastewater samples. In this context, we propose an alternative unsupervised method for clustering mutations according to their frequency trajectory over time and estimating group fitness from time series of pooled mutation prevalence data. Our model is a mixture of observed count data and latent group assignment and we use the expectation-maximization algorithm for model selection and parameter estimation. The application of our method to time series of SARS-CoV-2 sequencing data collected from wastewater treatment plants in France from October 2020 to April 2021 shows its ability to agnostically group mutations according to their probability of belonging to B.1.160, Alpha, Beta, B.1.177 variants with selection coefficient estimates per group in coherence with the viral dynamics in France reported by Nextstrain. Moreover, our method detected the Alpha variant as threatening as early as supervised methods (which track specific mutations over time) with the noticeable difference that, since unsupervised, it does not require any prior information on the set of mutations.
Abstract: Background: Understanding cardiovascular artery disease risk factors, the leading global cause of mortality, is crucial for influencing its etiology, prevalence, and treatment. This study aims to evaluate prognostic markers for coronary artery disease in Mashhad using Naive Bayes, REP Tree, J48, CART, and CHAID algorithms. Methods: Using data from the 2009 MASHAD STUDY, prognostic factors for coronary artery disease were determined with Naive Bayes, REP Tree, J48, CART, CHAID, and Random Forest algorithms using R 3.5.3 and WEKA 3.9.4. Model efficiency was compared by sensitivity, specificity, and accuracy. Cases were patients with coronary artery disease; each had three controls (totally 940). Results: Prognostic factors for coronary artery disease in Mashhad residents varied by algorithm. CHAID identified age, myocardial infarction history, and hypertension. CART included depression score and physical activity. REP added education level and anxiety score. NB included diabetes and family history. J48 highlighted father's heart disease and weight loss. CHAID had the highest accuracy (0.80). Conclusion: Key prognostic factors for coronary artery disease in CART and CHAID models include age, myocardial infarction history, hypertension, depression score, physical activity, and BMI. NB, REP Tree, and J48 identified numerous factors. CHAID had the highest accuracy, sensitivity, and specificity. CART offers simpler interpretation, aiding physician and paramedic model selection based on specific. Keywords: RF, Na\"ive Bayes, REP, J48 algorithms, Coronary Artery Disease (CAD).
The number of people living in this agricultural nation of ours, which is surrounded by lush greenery, is growing on a daily basis. As a result of this, the level of arable land is decreasing, as well as residential houses and industrial factories. The food crisis is becoming the main threat for us in the upcoming days. Because on the one hand, the population is increasing, and on the other hand, the amount of food crop production is decreasing due to the attack of diseases. Rice is one of the most significant cultivated crops since it provides food for more than half of the world's population. Bangladesh is dependent on rice (Oryza sativa) as a vital crop for its agriculture, but it faces a significant problem as a result of the ongoing decline in rice yield brought on by common diseases. Early disease detection is the main difficulty in rice crop cultivation. In this paper, we proposed our own dataset, which was collected from the Bangladesh field, and also applied deep learning and transfer learning models for the evaluation of the datasets. We elaborately explain our dataset and also give direction for further research work to serve society using this dataset. We applied a light CNN model and pre-trained InceptionNet-V2, EfficientNet-V2, and MobileNet-V2 models, which achieved 91.5% performance for the EfficientNet-V2 model of this work. The results obtained assaulted other models and even exceeded approaches that are considered to be part of the state of the art. It has been demonstrated by this study that it is possible to precisely and effectively identify diseases that affect rice leaves using this unbiased datasets. After analysis of the performance of different models, the proposed datasets are significant for the society for research work to provide solutions for decreasing rice leaf disease.
We describe two Monte Carlo algorithms for sampling from the integrated posterior distributions of a range of Bayesian mixture models. Both algorithms allow us to directly sample not only the assignment of observations to components but also the number of components, thereby fitting the model and performing model selection over the number of components in a single computation. The first algorithm is a traditional collapsed Gibbs sampler, albeit with an unusual move-set; the second builds on the first, adding rejection-free sampling from the prior over component assignments, to create an algorithm that has excellent mixing time in typical applications and outperforms current state-of-the-art methods, in some cases by a wide margin. We demonstrate our methods with a selection of applications to latent class analysis.
Whenever data-based systems are employed in engineering applications, defining an optimal statistical representation is subject to the problem of model selection. This paper focusses on how well models can generalise in Structural Health Monitoring (SHM). Although statistical model validation in this field is often performed heuristically, it is possible to estimate generalisation more rigorously using the bounds provided by Statistical Learning Theory (SLT). Therefore, this paper explores the selection process of a kernel smoother for modelling the impulse response of a linear oscillator from the perspective of SLT. It is demonstrated that incorporating domain knowledge into the regression problem yields a lower guaranteed risk, thereby enhancing generalisation.
Estimating long-term causal effects from short-term data is essential for decision-making in healthcare, economics, and industry, where long-term follow-up is often infeasible. Markov Decision Processes (MDPs) offer a principled framework for modeling outcomes as sequences of states, actions, and rewards over time. We introduce a semiparametric extension of Double Reinforcement Learning (DRL) for statistically efficient, model-robust inference on linear functionals of the Q-function, such as policy values, in infinite-horizon, time-homogeneous MDPs. By imposing semiparametric structure on the Q-function, our method relaxes the strong state overlap assumptions required by fully nonparametric approaches, improving efficiency and stability. To address computational and robustness challenges of minimax nuisance estimation, we develop a novel debiased plug-in estimator based on isotonic Bellman calibration, which integrates fitted Q-iteration with an isotonic regression step. This procedure leverages the Q-function as a data-driven dimension reduction, debiases all linear functionals of interest simultaneously, and enables nonparametric inference without explicit nuisance function estimation. Bellman calibration generalizes isotonic calibration to MDPs and may be of independent interest for prediction in reinforcement learning. Finally, we show that model selection for the Q-function incurs only second-order bias and extend the adaptive debiased machine learning (ADML) framework to MDPs for data-driven learning of semiparametric structure.
Mental health disorders pose a growing public health concern in the Arab world, emphasizing the need for accessible diagnostic and intervention tools. Large language models (LLMs) offer a promising approach, but their application in Arabic contexts faces challenges including limited labeled datasets, linguistic complexity, and translation biases. This study comprehensively evaluates 8 LLMs, including general multi-lingual models, as well as bi-lingual ones, on diverse mental health datasets (such as AraDepSu, Dreaddit, MedMCQA), investigating the impact of prompt design, language configuration (native Arabic vs. translated English, and vice versa), and few-shot prompting on diagnostic performance. We find that prompt engineering significantly influences LLM scores mainly due to reduced instruction following, with our structured prompt outperforming a less structured variant on multi-class datasets, with an average difference of 14.5\%. While language influence on performance was modest, model selection proved crucial: Phi-3.5 MoE excelled in balanced accuracy, particularly for binary classification, while Mistral NeMo showed superior performance in mean absolute error for severity prediction tasks. Few-shot prompting consistently improved performance, with particularly substantial gains observed for GPT-4o Mini on multi-class classification, boosting accuracy by an average factor of 1.58. These findings underscore the importance of prompt optimization, multilingual analysis, and few-shot learning for developing culturally sensitive and effective LLM-based mental health tools for Arabic-speaking populations.
This research provides an in-depth evaluation of various machine learning models for energy forecasting, focusing on the unique challenges of seasonal variations in student residential settings. The study assesses the performance of baseline models, such as LSTM and GRU, alongside state-of-the-art forecasting methods, including Autoregressive Feedforward Neural Networks, Transformers, and hybrid approaches. Special attention is given to predicting energy consumption amidst challenges like seasonal patterns, vacations, meteorological changes, and irregular human activities that cause sudden fluctuations in usage. The findings reveal that no single model consistently outperforms others across all seasons, emphasizing the need for season-specific model selection or tailored designs. Notably, the proposed Hyper Network based LSTM and MiniAutoEncXGBoost models exhibit strong adaptability to seasonal variations, effectively capturing abrupt changes in energy consumption during summer months. This study advances the energy forecasting field by emphasizing the critical role of seasonal dynamics and model-specific behavior in achieving accurate predictions.
Understanding firm conduct is crucial for industrial organization and antitrust policy. In this article, we develop a testing procedure based on the Rivers and Vuong non-nested model selection framework. Unlike existing methods that require estimating the demand and supply system, our approach compares the model fit of two first-stage price regressions. Through an extensive Monte Carlo study, we demonstrate that our test performs comparably to, or outperforms, existing methods in detecting collusion across various collusive scenarios. The results are robust to model misspecification, alternative functional forms for instruments, and data limitations. By simplifying the diagnosis of firm behavior, our method provides an efficient tool for researchers and regulators to assess industry conduct. Additionally, our approach offers a practical guideline for enhancing the strength of BLP-style instruments in demand estimation: once collusion is detected, researchers are advised to incorporate the product characteristics of colluding partners into own-firm instruments while excluding them from other-firm instruments.