The preservation of soil health is a critical challenge in the 21st century due to its significant impact on agriculture, human health, and biodiversity. We provide the first deep investigation of the predictive potential of machine learning models to understand the connections between soil and biological phenotypes. We investigate an integrative framework performing accurate machine learning-based prediction of plant phenotypes from biological, chemical, and physical properties of the soil via two models: random forest and Bayesian neural network. We show that prediction is improved when incorporating environmental features like soil physicochemical properties and microbial population density into the models, in addition to the microbiome information. Exploring various data preprocessing strategies confirms the significant impact of human decisions on predictive performance. We show that the naive total sum scaling normalization that is commonly used in microbiome research is not the optimal strategy to maximize predictive power. Also, we find that accurately defined labels are more important than normalization, taxonomic level or model characteristics. In cases where humans are unable to classify samples accurately, machine learning model performance is limited. Lastly, we provide domain scientists via a full model selection decision tree to identify the human choices that optimize model prediction power. Our work is accompanied by open source reproducible scripts (https://github.com/solislemuslab/soil-microbiome-nn) for maximum outreach among the microbiome research community.
Selecting hyperparameters in deep learning greatly impacts its effectiveness but requires manual effort and expertise. Recent works show that Bayesian model selection with Laplace approximations can allow to optimize such hyperparameters just like standard neural network parameters using gradients and on the training data. However, estimating a single hyperparameter gradient requires a pass through the entire dataset, limiting the scalability of such algorithms. In this work, we overcome this issue by introducing lower bounds to the linearized Laplace approximation of the marginal likelihood. In contrast to previous estimators, these bounds are amenable to stochastic-gradient-based optimization and allow to trade off estimation accuracy against computational complexity. We derive them using the function-space form of the linearized Laplace, which can be estimated using the neural tangent kernel. Experimentally, we show that the estimators can significantly accelerate gradient-based hyperparameter optimization.
Test-Time Adaptation (TTA) has recently emerged as a promising approach for tackling the robustness challenge under distribution shifts. However, the lack of consistent settings and systematic studies in prior literature hinders thorough assessments of existing methods. To address this issue, we present TTAB, a test-time adaptation benchmark that encompasses ten state-of-the-art algorithms, a diverse array of distribution shifts, and two evaluation protocols. Through extensive experiments, our benchmark reveals three common pitfalls in prior efforts. First, selecting appropriate hyper-parameters, especially for model selection, is exceedingly difficult due to online batch dependency. Second, the effectiveness of TTA varies greatly depending on the quality and properties of the model being adapted. Third, even under optimal algorithmic conditions, none of the existing methods are capable of addressing all common types of distribution shifts. Our findings underscore the need for future research in the field to conduct rigorous evaluations on a broader set of models and shifts, and to re-examine the assumptions behind the empirical success of TTA. Our code is available at \url{https://github.com/lins-lab/ttab}.
Automated model selection is an important application in science and engineering. In this work, we develop a learning approach for identifying structured dynamical systems from undersampled and noisy spatiotemporal data. The learning is performed by a sparse least-squares fitting over a large set of candidate functions via a nonconvex $\ell_1-\ell_2$ sparse optimization solved by the alternating direction method of multipliers. Using a Bernstein-like inequality with a coherence condition, we show that if the set of candidate functions forms a structured random sampling matrix of a bounded orthogonal system, the recovery is stable and the error is bounded. The learning approach is validated on synthetic data generated by the viscous Burgers' equation and two reaction-diffusion equations. The computational results demonstrate the theoretical guarantees of success and the efficiency with respect to the ambient dimension and the number of candidate functions.
Massively multilingual language models have displayed strong performance in zero-shot (ZS-XLT) and few-shot (FS-XLT) cross-lingual transfer setups, where models fine-tuned on task data in a source language are transferred without any or with only a few annotated instances to the target language(s). However, current work typically overestimates model performance as fine-tuned models are frequently evaluated at model checkpoints that generalize best to validation instances in the target languages. This effectively violates the main assumptions of "true" ZS-XLT and FS-XLT. Such XLT setups require robust methods that do not depend on labeled target language data for validation and model selection. In this work, aiming to improve the robustness of "true" ZS-XLT and FS-XLT, we propose a simple and effective method that averages different checkpoints (i.e., model snapshots) during task fine-tuning. We conduct exhaustive ZS-XLT and FS-XLT experiments across higher-level semantic tasks (NLI, extractive QA) and lower-level token classification tasks (NER, POS). The results indicate that averaging model checkpoints yields systematic and consistent performance gains across diverse target languages in all tasks. Importantly, it simultaneously substantially desensitizes XLT to varying hyperparameter choices in the absence of target language validation. We also show that checkpoint averaging benefits performance when further combined with run averaging (i.e., averaging the parameters of models fine-tuned over independent runs).
Domain generalization aims to solve the challenge of Out-of-Distribution (OOD) generalization by leveraging common knowledge learned from multiple training domains to generalize to unseen test domains. To accurately evaluate the OOD generalization ability, it is required that test data information is unavailable. However, the current domain generalization protocol may still have potential test data information leakage. This paper examines the risks of test data information leakage from two aspects of the current evaluation protocol: supervised pretraining on ImageNet and oracle model selection. We propose modifications to the current protocol that we should employ self-supervised pretraining or train from scratch instead of employing the current supervised pretraining, and we should use multiple test domains. These would result in a more precise evaluation of OOD generalization ability. We also rerun the algorithms with the modified protocol and introduce new leaderboards to encourage future research in domain generalization with a fairer comparison.
Contextual Bayesian Optimization (CBO) efficiently optimizes black-box functions with respect to design variables, while simultaneously integrating contextual information regarding the environment, such as experimental conditions. However, the relevance of contextual variables is not necessarily known beforehand. Moreover, contextual variables can sometimes be optimized themselves at an additional cost, a setting overlooked by current CBO algorithms. Cost-sensitive CBO would simply include optimizable contextual variables as part of the design variables based on their cost. Instead, we adaptively select a subset of contextual variables to include in the optimization, based on the trade-off between their relevance and the additional cost incurred by optimizing them compared to leaving them to be determined by the environment. We learn the relevance of contextual variables by sensitivity analysis of the posterior surrogate model while minimizing the cost of optimization by leveraging recent developments on early stopping for BO. We empirically evaluate our proposed Sensitivity-Analysis-Driven Contextual BO (SADCBO) method against alternatives on both synthetic and real-world experiments, together with extensive ablation studies, and demonstrate a consistent improvement across examples.
Chain-of-Thought (CoT) and Program-Aided Language Models (PAL) represent two distinct reasoning methods, each with its own strengths. CoT employs natural language, offering flexibility and interpretability, while PAL utilizes programming language, yielding more structured and rigorous logic. We introduce a model selection method to combine the best of both worlds by employing a large language model (LLM) to dynamically select between them. Our theoretical analysis underscores the feasibility of this method, which is further corroborated by empirical results. Our proposed method demonstrates significant performance improvements across eight reasoning datasets with Codex, ChatGPT, and GPT-4. Additionally, our method is complementary to self-consistency; when integrated, it can further enhance performance while significantly reducing computation costs. Moreover, we achieve new state-of-the-art results on GSM8K and SVAMP, with respective accuracies of 96.8% and 93.7%. Our code, data and prompts are available at https://github.com/XuZhao0/Model-Selection-Reasoning
Classification model selection is a process of identifying a suitable model class for a given classification task on a dataset. Traditionally, model selection is based on cross-validation, meta-learning, and user preferences, which are often time-consuming and resource-intensive. The performance of any machine learning classification task depends on the choice of the model class, the learning algorithm, and the dataset's characteristics. Our work proposes a novel method for automatic classification model selection from a set of candidate model classes by determining the empirical model-fitness for a dataset based only on its clustering indices. Clustering Indices measure the ability of a clustering algorithm to induce good quality neighborhoods with similar data characteristics. We propose a regression task for a given model class, where the clustering indices of a given dataset form the features and the dependent variable represents the expected classification performance. We compute the dataset clustering indices and directly predict the expected classification performance using the learned regressor for each candidate model class to recommend a suitable model class for dataset classification. We evaluate our model selection method through cross-validation with 60 publicly available binary class datasets and show that our top3 model recommendation is accurate for over 45 of 60 datasets. We also propose an end-to-end Automated ML system for data classification based on our model selection method. We evaluate our end-to-end system against popular commercial and noncommercial Automated ML systems using a different collection of 25 public domain binary class datasets. We show that the proposed system outperforms other methods with an excellent average rank of 1.68.
We present the analysis of the topological graph descriptor Local Degree Profile (LDP), which forms a widely used structural baseline for graph classification. Our study focuses on model evaluation in the context of the recently developed fair evaluation framework, which defines rigorous routines for model selection and evaluation for graph classification, ensuring reproducibility and comparability of the results. Based on the obtained insights, we propose a new baseline algorithm called Local Topological Profile (LTP), which extends LDP by using additional centrality measures and local vertex descriptors. The new approach provides the results outperforming or very close to the latest GNNs for all datasets used. Specifically, state-of-the-art results were obtained for 4 out of 9 benchmark datasets. We also consider computational aspects of LDP-based feature extraction and model construction to propose practical improvements affecting execution speed and scalability. This allows for handling modern, large datasets and extends the portfolio of benchmarks used in graph representation learning. As the outcome of our work, we obtained LTP as a simple to understand, fast and scalable, still robust baseline, capable of outcompeting modern graph classification models such as Graph Isomorphism Network (GIN). We provide open-source implementation at \href{https://github.com/j-adamczyk/LTP}{GitHub}.
Recommender Systems today are still mostly evaluated in terms of accuracy, with other aspects beyond the immediate relevance of recommendations, such as diversity, long-term user retention and fairness, often taking a back seat. Moreover, reconciling multiple performance perspectives is by definition indeterminate, presenting a stumbling block to those in the pursuit of rounded evaluation of Recommender Systems. EvalRS 2022 -- a data challenge designed around Multi-Objective Evaluation -- was a first practical endeavour, providing many insights into the requirements and challenges of balancing multiple objectives in evaluation. In this work, we reflect on EvalRS 2022 and expound upon crucial learnings to formulate a first-principles approach toward Multi-Objective model selection, and outline a set of guidelines for carrying out a Multi-Objective Evaluation challenge, with potential applicability to the problem of rounded evaluation of competing models in real-world deployments.
A well-known problem when learning from user clicks are inherent biases prevalent in the data, such as position or trust bias. Click models are a common method for extracting information from user clicks, such as document relevance in web search, or to estimate click biases for downstream applications such as counterfactual learning-to-rank, ad placement, or fair ranking. Recent work shows that the current evaluation practices in the community fail to guarantee that a well-performing click model generalizes well to downstream tasks in which the ranking distribution differs from the training distribution, i.e., under covariate shift. In this work, we propose an evaluation metric based on conditional independence testing to detect a lack of robustness to covariate shift in click models. We introduce the concept of debiasedness in click modeling and derive a metric for measuring it. In extensive semi-synthetic experiments, we show that our proposed metric helps to predict the downstream performance of click models under covariate shift and is useful in an off-policy model selection setting.
Wind turbine power curve models translate ambient conditions into turbine power output. They are essential for energy yield prediction and turbine performance monitoring. In recent years, increasingly complex machine learning methods have become state-of-the-art for this task. Nevertheless, they frequently encounter criticism due to their apparent lack of transparency, which raises concerns regarding their performance in non-stationary environments, such as those faced by wind turbines. We, therefore, introduce an explainable artificial intelligence (XAI) framework to investigate and validate strategies learned by data-driven power curve models from operational wind turbine data. With the help of simple, physics-informed baseline models it enables an automated evaluation of machine learning models beyond standard error metrics. Alongside this novel tool, we present its efficacy for a more informed model selection. We show, for instance, that learned strategies can be meaningful indicators for a model's generalization ability in addition to test set errors, especially when only little data is available. Moreover, the approach facilitates an understanding of how decisions along the machine learning pipeline, such as data selection, pre-processing, or training parameters, affect learned strategies. In a practical example, we demonstrate the framework's utilisation to obtain more physically meaningful models, a prerequisite not only for robustness but also for insights into turbine operation by domain experts. The latter, we demonstrate in the context of wind turbine performance monitoring. Alongside this paper, we publish a Python implementation of the presented framework and hope this can guide researchers and practitioners alike toward training, selecting and utilizing more transparent and robust data-driven wind turbine power curve models.
When choosing between competing symbolic models for a data set, a human will naturally prefer the "simpler" expression or the one which more closely resembles equations previously seen in a similar context. This suggests a non-uniform prior on functions, which is, however, rarely considered within a symbolic regression (SR) framework. In this paper we develop methods to incorporate detailed prior information on both functions and their parameters into SR. Our prior on the structure of a function is based on a $n$-gram language model, which is sensitive to the arrangement of operators relative to one another in addition to the frequency of occurrence of each operator. We also develop a formalism based on the Fractional Bayes Factor to treat numerical parameter priors in such a way that models may be fairly compared though the Bayesian evidence, and explicitly compare Bayesian, Minimum Description Length and heuristic methods for model selection. We demonstrate the performance of our priors relative to literature standards on benchmarks and a real-world dataset from the field of cosmology.
We introduce a new dataset named WikiVitals which contains a large graph of 48k mutually referred Wikipedia articles classified into 32 categories and connected by 2.3M edges. Our aim is to rigorously evaluate the contributions of three distinct sources of information to the label prediction in a semi-supervised node classification setting, namely the content of the articles, their connections with each other and the correlations among their labels. We perform this evaluation using a Graph Markov Neural Network which provides a theoretically principled model for this task and we conduct a detailed evaluation of the contributions of each sources of information using a clear separation of model selection and model assessment. One interesting observation is that including the effect of label dependencies is more relevant for sparse train sets than it is for dense train sets.
Domain generalization aims to learn a model with good generalization ability, that is, the learned model should not only perform well on several seen domains but also on unseen domains with different data distributions. State-of-the-art domain generalization methods typically train a representation function followed by a classifier jointly to minimize both the classification risk and the domain discrepancy. However, when it comes to model selection, most of these methods rely on traditional validation routines that select models solely based on the lowest classification risk on the validation set. In this paper, we theoretically demonstrate a trade-off between minimizing classification risk and mitigating domain discrepancy, i.e., it is impossible to achieve the minimum of these two objectives simultaneously. Motivated by this theoretical result, we propose a novel model selection method suggesting that the validation process should account for both the classification risk and the domain discrepancy. We validate the effectiveness of the proposed method by numerical results on several domain generalization datasets.
In this paper, we introduce a new approach to address the challenge of generalization in hyperspectral anomaly detection (AD). Our method eliminates the need for adjusting parameters or retraining on new test scenes as required by most existing methods. Employing an image-level training paradigm, we achieve a general anomaly enhancement network for hyperspectral AD that only needs to be trained once. Trained on a set of anomaly-free hyperspectral images with random masks, our network can learn the spatial context characteristics between anomalies and background in an unsupervised way. Additionally, a plug-and-play model selection module is proposed to search for a spatial-spectral transform domain that is more suitable for AD task than the original data. To establish a unified benchmark to comprehensively evaluate our method and existing methods, we develop a large-scale hyperspectral AD dataset (HAD100) that includes 100 real test scenes with diverse anomaly targets. In comparison experiments, we combine our network with a parameter-free detector and achieve the optimal balance between detection accuracy and inference speed among state-of-the-art AD methods. Experimental results also show that our method still achieves competitive performance when the training and test set are captured by different sensor devices. Our code is available at https://github.com/ZhaoxuLi123/AETNet.
PyVBMC is a Python implementation of the Variational Bayesian Monte Carlo (VBMC) algorithm for posterior and model inference for black-box computational models (Acerbi, 2018, 2020). VBMC is an approximate inference method designed for efficient parameter estimation and model assessment when model evaluations are mildly-to-very expensive (e.g., a second or more) and/or noisy. Specifically, VBMC computes: - a flexible (non-Gaussian) approximate posterior distribution of the model parameters, from which statistics and posterior samples can be easily extracted; - an approximation of the model evidence or marginal likelihood, a metric used for Bayesian model selection. PyVBMC can be applied to any computational or statistical model with up to roughly 10-15 continuous parameters, with the only requirement that the user can provide a Python function that computes the target log likelihood of the model, or an approximation thereof (e.g., an estimate of the likelihood obtained via simulation or Monte Carlo methods). PyVBMC is particularly effective when the model takes more than about a second per evaluation, with dramatic speed-ups of 1-2 orders of magnitude when compared to traditional approximate inference methods. Extensive benchmarks on both artificial test problems and a large number of real models from the computational sciences, particularly computational and cognitive neuroscience, show that VBMC generally - and often vastly - outperforms alternative methods for sample-efficient Bayesian inference, and is applicable to both exact and simulator-based models (Acerbi, 2018, 2019, 2020). PyVBMC brings this state-of-the-art inference algorithm to Python, along with an easy-to-use Pythonic interface for running the algorithm and manipulating and visualizing its results.
Sports betting's recent federal legalisation in the USA coincides with the golden age of machine learning. If bettors can leverage data to reliably predict the probability of an outcome, they can recognise when the bookmaker's odds are in their favour. As sports betting is a multi-billion dollar industry in the USA alone, identifying such opportunities could be extremely lucrative. Many researchers have applied machine learning to the sports outcome prediction problem, generally using accuracy to evaluate the performance of predictive models. We hypothesise that for the sports betting problem, model calibration is more important than accuracy. To test this hypothesis, we train models on NBA data over several seasons and run betting experiments on a single season, using published odds. We show that using calibration, rather than accuracy, as the basis for model selection leads to greater returns, on average (return on investment of $+34.69\%$ versus $-35.17\%$) and in the best case ($+36.93\%$ versus $+5.56\%$). These findings suggest that for sports betting (or any probabilistic decision-making problem), calibration is a more important metric than accuracy. Sports bettors who wish to increase profits should therefore select their predictive model based on calibration, rather than accuracy.
The problem of model selection with a limited number of experimental trials has received considerable attention in cognitive science, where the role of experiments is to discriminate between theories expressed as computational models. Research on this subject has mostly been restricted to optimal experiment design with analytically tractable models. However, cognitive models of increasing complexity, with intractable likelihoods, are becoming more commonplace. In this paper, we propose BOSMOS: an approach to experimental design that can select between computational models without tractable likelihoods. It does so in a data-efficient manner, by sequentially and adaptively generating informative experiments. In contrast to previous approaches, we introduce a novel simulator-based utility objective for design selection, and a new approximation of the model likelihood for model selection. In simulated experiments, we demonstrate that the proposed BOSMOS technique can accurately select models in up to 2 orders of magnitude less time than existing LFI alternatives for three cognitive science tasks: memory retention, sequential signal detection and risky choice.