Deep Neural Networks (DNNs) are increasingly deployed across applications. However, ensuring their reliability remains a challenge, and in many situations, alternative models with similar functionality and accuracy are available. Traditional accuracy-based evaluations often fail to capture behavioral differences between models, especially with limited test datasets, making it difficult to select or combine models effectively. Differential testing addresses this by generating test inputs that expose discrepancies in DNN model behavior. However, existing approaches face significant limitations: many rely on model internals or are constrained by available seed inputs. To address these challenges, we propose DiffGAN, a black-box test image generation approach for differential testing of DNN models. DiffGAN leverages a Generative Adversarial Network (GAN) and the Non-dominated Sorting Genetic Algorithm II to generate diverse and valid triggering inputs that reveal behavioral discrepancies between models. DiffGAN employs two custom fitness functions, focusing on diversity and divergence, to guide the exploration of the GAN input space and identify discrepancies between models' outputs. By strategically searching this space, DiffGAN generates inputs with specific features that trigger differences in model behavior. DiffGAN is black-box, making it applicable in more situations. We evaluate DiffGAN on eight DNN model pairs trained on widely used image datasets. Our results show DiffGAN significantly outperforms a SOTA baseline, generating four times more triggering inputs, with greater diversity and validity, within the same budget. Additionally, the generated inputs improve the accuracy of a machine learning-based model selection mechanism, which selects the best-performing model based on input characteristics and can serve as a smart output voting mechanism when using alternative models.
The analysis of neural power spectra plays a crucial role in understanding brain function and dysfunction. While recent efforts have led to the development of methods for decomposing spectral data, challenges remain in performing statistical analysis and group-level comparisons. Here, we introduce Bayesian Spectral Decomposition (BSD), a Bayesian framework for analysing neural spectral power. BSD allows for the specification, inversion, comparison, and analysis of parametric models of neural spectra, addressing limitations of existing methods. We first establish the face validity of BSD on simulated data and show how it outperforms an established method (\fooof{}) for peak detection on artificial spectral data. We then demonstrate the efficacy of BSD on a group-level study of EEG spectra in 204 healthy subjects from the LEMON dataset. Our results not only highlight the effectiveness of BSD in model selection and parameter estimation, but also illustrate how BSD enables straightforward group-level regression of the effect of continuous covariates such as age. By using Bayesian inference techniques, BSD provides a robust framework for studying neural spectral data and their relationship to brain function and dysfunction.
We present an accelerated pipeline, based on high-performance computing techniques and normalizing flows, for joint Bayesian parameter estimation and model selection and demonstrate its efficiency in gravitational wave astrophysics. We integrate the Jim inference toolkit, a normalizing flow-enhanced Markov chain Monte Carlo (MCMC) sampler, with the learned harmonic mean estimator. Our Bayesian evidence estimates run on $1$ GPU are consistent with traditional nested sampling techniques run on $16$ CPU cores, while reducing the computation time by factors of $5\times$ and $15\times$ for $4$-dimensional and $11$-dimensional gravitational wave inference problems, respectively. Our code is available in well-tested and thoroughly documented open-source packages, ensuring accessibility and reproducibility for the wider research community.
A number of popular transfer learning methods rely on grid search to select regularization hyperparameters that control over-fitting. This grid search requirement has several key disadvantages: the search is computationally expensive, requires carving out a validation set that reduces the size of available data for model training, and requires practitioners to specify candidate values. In this paper, we propose an alternative to grid search: directly learning regularization hyperparameters on the full training set via model selection techniques based on the evidence lower bound ("ELBo") objective from variational methods. For deep neural networks with millions of parameters, we specifically recommend a modified ELBo that upweights the influence of the data likelihood relative to the prior while remaining a valid bound on the evidence for Bayesian model selection. Our proposed technique overcomes all three disadvantages of grid search. We demonstrate effectiveness on image classification tasks on several datasets, yielding heldout accuracy comparable to existing approaches with far less compute time.
Model selection is the process of choosing from a class of candidate models given data. For instance, methods such as the LASSO and sparse identification of nonlinear dynamics (SINDy) formulate model selection as finding a sparse solution to a linear system of equations determined by training data. However, absent strong assumptions, such methods are highly unstable: if a single data point is removed from the training set, a different model may be selected. In this paper, we present a new approach to stabilizing model selection with theoretical stability guarantees that leverages a combination of bagging and an ''inflated'' argmax operation. Our method selects a small collection of models that all fit the data, and it is stable in that, with high probability, the removal of any training point will result in a collection of selected models that overlaps with the original collection. We illustrate this method in (a) a simulation in which strongly correlated covariates make standard LASSO model selection highly unstable, (b) a Lotka-Volterra model selection problem focused on identifying how competition in an ecosystem influences species' abundances, and (c) a graph subset selection problem using cell-signaling data from proteomics. In these settings, the proposed method yields stable, compact, and accurate collections of selected models, outperforming a variety of benchmarks.
Open-ended text generation has become a prominent task in natural language processing due to the rise of powerful (large) language models. However, evaluating the quality of these models and the employed decoding strategies remains challenging because of trade-offs among widely used metrics such as coherence, diversity, and perplexity. Decoding methods often excel in some metrics while underperforming in others, complicating the establishment of a clear ranking. In this paper, we present novel ranking strategies within this multicriteria framework. Specifically, we employ benchmarking approaches based on partial orderings and present a new summary metric designed to balance existing automatic indicators, providing a more holistic evaluation of text generation quality. Furthermore, we discuss the alignment of these approaches with human judgments. Our experiments demonstrate that the proposed methods offer a robust way to compare decoding strategies, exhibit similarities with human preferences, and serve as valuable tools in guiding model selection for open-ended text generation tasks. Finally, we suggest future directions for improving evaluation methodologies in text generation. Our codebase, datasets, and models are publicly available.
Quantifying robustness in a single measure for the purposes of model selection, development of adversarial training methods, and anticipating trends has so far been elusive. The simplest metric to consider is the number of trainable parameters in a model but this has previously been shown to be insufficient at explaining robustness properties. A variety of other metrics, such as ones based on boundary thickness and gradient flatness have been proposed but have been shown to be inadequate proxies for robustness. In this work, we investigate the relationship between a model's effective dimensionality, which can be thought of as model complexity, and its robustness properties. We run experiments on commercial-scale models that are often used in real-world environments such as YOLO and ResNet. We reveal a near-linear inverse relationship between effective dimensionality and adversarial robustness, that is models with a lower dimensionality exhibit better robustness. We investigate the effect of a variety of adversarial training methods on effective dimensionality and find the same inverse linear relationship present, suggesting that effective dimensionality can serve as a useful criterion for model selection and robustness evaluation, providing a more nuanced and effective metric than parameter count or previously-tested measures.
Even with the significant advances of AlphaFold-Multimer (AF-Multimer) and AlphaFold3 (AF3) in protein complex structure prediction, their accuracy is still not comparable with monomer structure prediction. Efficient quality assessment (QA) or estimation of model accuracy (EMA) models that can evaluate the quality of the predicted protein-complexes without knowing their native structures, are of key importance for protein structure generation and model selection. In this paper, we leverage persistent homology (PH) to capture the atomic-level topological information around residues and design a topological deep learning-based QA method, TopoQA, to assess the accuracy of protein complex interfaces. We integrate PH from topological data analysis into graph neural networks (GNNs) to characterize complex higher-order structures that GNNs might overlook, enhancing the learning of the relationship between the topological structure of complex interfaces and quality scores. Our TopoQA model is extensively validated based on the two most-widely used datasets, DBM55-AF2 and HAF2, along with our newly constructed ABAG-AF3 dataset to facilitate comparisons with AF3. For all three datasets, TopoQA outperforms AF-Multimer-based AF2Rank and shows an advantage over AF3 in nearly half of the targets. In particular, in the DBM55-AF2 dataset, a ranking loss of 73.6% lower than AF-Multimer-based AF2Rank is obtained. Further, other than AF-Multimer and AF3, we have also extensively compared with nearly-all the state-of-the-art models (as far as we know), it has been found that our TopoQA can achieve the highest Top 10 Hit-rate on the DBM55-AF2 dataset and the lowest ranking loss on the HAF2 dataset. Ablation experiments show that our topological features significantly improve the model performance. At the same time, our method also provides a new paradigm for protein structure representation learning.
With the growing number of forecasting techniques and the increasing significance of forecast-based operation - particularly in the rapidly evolving energy sector - selecting the most effective forecasting model has become a critical task. Given the dynamic nature of energy forecasting, it is highly advantageous to assess the superiority of forecasting models not only retrospectively but continuously in real-time as new data and evidence becomes available, while simultaneously providing strong probabilistic guarantees for these decisions. In this work, we show that this can be achieved through the mathematical concept of e-values, which has recently gained massive attention in the field of statistics. It allows for unified construction principles for powerful tests and accurate statistical decisions, which can be evaluated at any chosen time points while maintaining an overall probabilistic error control. We extend the use of e-values by developing a simple persistence approach that dynamically combines input forecasts to generate new fused predictions. To demonstrate the performance of our method we apply it to electricity demand forecasts based on different artificial intelligence based models. Our results indicate that e-values are able to improve the accuracy and reliability of forecasts in a dynamic environment, offering a valuable tool for real-time decision-making in the energy sector.
Automated Machine Learning (AutoML) approaches encompass traditional methods that optimize fixed pipelines for model selection and ensembling, as well as newer LLM-based frameworks that autonomously build pipelines. While LLM-based agents have shown promise in automating machine learning tasks, they often generate low-diversity and suboptimal code, even after multiple iterations. To overcome these limitations, we introduce Tree-Search Enhanced LLM Agents (SELA), an innovative agent-based system that leverages Monte Carlo Tree Search (MCTS) to optimize the AutoML process. By representing pipeline configurations as trees, our framework enables agents to conduct experiments intelligently and iteratively refine their strategies, facilitating a more effective exploration of the machine learning solution space. This novel approach allows SELA to discover optimal pathways based on experimental feedback, improving the overall quality of the solutions. In an extensive evaluation across 20 machine learning datasets, we compare the performance of traditional and agent-based AutoML methods, demonstrating that SELA achieves a win rate of 65% to 80% against each baseline across all datasets. These results underscore the significant potential of agent-based strategies in AutoML, offering a fresh perspective on tackling complex machine learning challenges.
Function-space priors in Bayesian Neural Networks provide a more intuitive approach to embedding beliefs directly into the model's output, thereby enhancing regularization, uncertainty quantification, and risk-aware decision-making. However, imposing function-space priors on BNNs is challenging. We address this task through optimization techniques that explore how trainable activations can accommodate complex priors and match intricate target function distributions. We discuss critical learning challenges, including identifiability, loss construction, and symmetries that arise in this context. Furthermore, we enable evidence maximization to facilitate model selection by conditioning the functional priors on additional hyperparameters. Our empirical findings demonstrate that even BNNs with a single wide hidden layer, when equipped with these adaptive trainable activations and conditioning strategies, can effectively achieve high-fidelity function-space priors, providing a robust and flexible framework for enhancing Bayesian neural network performance.
Unsupervised validation of anomaly-detection models is a highly challenging task. While the common practices for model validation involve a labeled validation set, such validation sets cannot be constructed when the underlying datasets are unlabeled. The lack of robust and efficient unsupervised model-validation techniques presents an acute challenge in the implementation of automated anomaly-detection pipelines, especially when there exists no prior knowledge of the model's performance on similar datasets. This work presents a new paradigm to automated validation of anomaly-detection models, inspired by real-world, collaborative decision-making mechanisms. We focus on two commonly-used, unsupervised model-validation tasks -- model selection and model evaluation -- and provide extensive experimental results that demonstrate the accuracy and robustness of our approach on both tasks.
In this paper, we make a case for a proxy for large language models which has explicit support for cost-saving optimizations. We design LLMProxy, which supports three key optimizations: model selection, context management, and caching. These optimizations present tradeoffs in terms of cost, inference time, and response quality, which applications can navigate through our high level, bidirectional interface. As a case study, we implement a WhatsApp-based Q&A service that uses LLMProxy to provide a rich set of features to the users. This service is deployed on a small scale (100+ users) leveraging the cloud; it has been operational for 15+ weeks and users have asked 1400+ questions so far. We report on the experiences of running this service as well as microbenchmark the specific benefits of the various cost-optimizations we present in this paper.
Pre-trained transformer models have shown great promise in various natural language processing tasks, including personalized news recommendations. To harness the power of these models, we introduce Transformers4NewsRec, a new Python framework built on the \textbf{Transformers} library. This framework is designed to unify and compare the performance of various news recommendation models, including deep neural networks and graph-based models. Transformers4NewsRec offers flexibility in terms of model selection, data preprocessing, and evaluation, allowing both quantitative and qualitative analysis.
Automated Machine Learning (AutoML) has simplified complex ML processes such as data pre-processing, model selection, and hyper-parameter searching. However, traditional AutoML frameworks focus solely on discriminative tasks, often falling short in tackling AutoML for generative models. Additionally, these frameworks lack interpretability and user engagement during the training process, primarily due to the absence of human-centered design. It leads to a lack of transparency in final decision-making and limited user control, potentially reducing trust and adoption of AutoML methods. To address these limitations, we introduce UniAutoML, a human-centered AutoML framework that leverages Large Language Models (LLMs) to unify AutoML for both discriminative (e.g., Transformers and CNNs for classification or regression tasks) and generative tasks (e.g., fine-tuning diffusion models or LLMs). The human-centered design of UniAutoML innovatively features a conversational user interface (CUI) that facilitates natural language interactions, providing users with real-time guidance, feedback, and progress updates for better interpretability. This design enhances transparency and user control throughout the AutoML training process, allowing users to seamlessly break down or modify the model being trained. To mitigate potential risks associated with LLM generated content, UniAutoML incorporates a safety guardline that filters inputs and censors outputs. We evaluated UniAutoML's performance and usability through experiments on eight diverse datasets and user studies involving 25 participants, demonstrating that UniAutoML not only enhances performance but also improves user control and trust. Our human-centered design bridges the gap between AutoML capabilities and user understanding, making ML more accessible to a broader audience.
Fairness-aware Machine Learning (FairML) applications are often characterized by complex social objectives and legal requirements, frequently involving multiple, potentially conflicting notions of fairness. Despite the well-known Impossibility Theorem of Fairness and extensive theoretical research on the statistical and socio-technical trade-offs between fairness metrics, many FairML tools still optimize or constrain for a single fairness objective. However, this one-sided optimization can inadvertently lead to violations of other relevant notions of fairness. In this socio-technical and empirical study, we frame fairness as a many-objective (MaO) problem by treating fairness metrics as conflicting objectives. We introduce ManyFairHPO, a human-in-the-loop, fairness-aware model selection framework that enables practitioners to effectively navigate complex and nuanced fairness objective landscapes. ManyFairHPO aids in the identification, evaluation, and balancing of fairness metric conflicts and their related social consequences, leading to more informed and socially responsible model-selection decisions. Through a comprehensive empirical evaluation and a case study on the Law School Admissions problem, we demonstrate the effectiveness of ManyFairHPO in balancing multiple fairness objectives, mitigating risks such as self-fulfilling prophecies, and providing interpretable insights to guide stakeholders in making fairness-aware modeling decisions.
We introduce MODEL SELECTOR, a framework for label-efficient selection of pretrained classifiers. Given a pool of unlabeled target data, MODEL SELECTOR samples a small subset of highly informative examples for labeling, in order to efficiently identify the best pretrained model for deployment on this target dataset. Through extensive experiments, we demonstrate that MODEL SELECTOR drastically reduces the need for labeled data while consistently picking the best or near-best performing model. Across 18 model collections on 16 different datasets, comprising over 1,500 pretrained models, MODEL SELECTOR reduces the labeling cost by up to 94.15% to identify the best model compared to the cost of the strongest baseline. Our results further highlight the robustness of MODEL SELECTOR in model selection, as it reduces the labeling cost by up to 72.41% when selecting a near-best model, whose accuracy is only within 1% of the best model.
Reward shaping is critical in reinforcement learning (RL), particularly for complex tasks where sparse rewards can hinder learning. However, choosing effective shaping rewards from a set of reward functions in a computationally efficient manner remains an open challenge. We propose Online Reward Selection and Policy Optimization (ORSO), a novel approach that frames the selection of shaping reward function as an online model selection problem. ORSO automatically identifies performant shaping reward functions without human intervention with provable regret guarantees. We demonstrate ORSO's effectiveness across various continuous control tasks. Compared to prior approaches, ORSO significantly reduces the amount of data required to evaluate a shaping reward function, resulting in superior data efficiency and a significant reduction in computational time (up to 8 times). ORSO consistently identifies high-quality reward functions outperforming prior methods by more than 50% and on average identifies policies as performant as the ones learned using manually engineered reward functions by domain experts.
The widespread applicability of large language models (LLMs) has increased the availability of many fine-tuned models of various sizes targeting specific tasks. Given a set of such specialized models, to maximize overall performance, it is important to figure out the optimal strategy for selecting the right model for a given user query. An effective strategy could drastically increase overall performance and even offer improvements over a single large monolithic model. Existing approaches typically fall into two categories: routing, where a single model is selected for each query, and cascading, which runs a sequence of increasingly larger models until a satisfactory answer is obtained. However, both have notable limitations: routing commits to an initial model without flexibility, while cascading requires executing every model in sequence, which can be inefficient. Additionally, the conditions under which these strategies are provably optimal remain unclear. In this work, we derive optimal strategies for both routing and cascading. Building on this analysis, we propose a novel approach called cascade routing, which combines the adaptability of routing with the cost-efficiency of cascading. Our experiments demonstrate that cascade routing consistently outperforms both routing and cascading across a variety of settings, improving both output quality and lowering computational cost, thus offering a unified and efficient solution to the model selection problem.
Machine learning (ML) has become a ubiquitous tool across various domains of data mining and big data analysis. The efficacy of ML models depends heavily on high-quality datasets, which are often complicated by the presence of missing values. Consequently, the performance and generalization of ML models are at risk in the face of such datasets. This paper aims to examine the nuanced impact of missing values on ML workflows, including their types, causes, and consequences. Our analysis focuses on the challenges posed by missing values, including biased inferences, reduced predictive power, and increased computational burdens. The paper further explores strategies for handling missing values, including imputation techniques and removal strategies, and investigates how missing values affect model evaluation metrics and introduces complexities in cross-validation and model selection. The study employs case studies and real-world examples to illustrate the practical implications of addressing missing values. Finally, the discussion extends to future research directions, emphasizing the need for handling missing values ethically and transparently. The primary goal of this paper is to provide insights into the pervasive impact of missing values on ML models and guide practitioners toward effective strategies for achieving robust and reliable model outcomes.