We propose an AutoML system that enables model selection on clustering problems by leveraging optimal transport-based dataset similarity. Our objective is to establish a comprehensive AutoML pipeline for clustering problems and provide recommendations for selecting the most suitable algorithms, thus opening up a new area of AutoML beyond the traditional supervised learning settings. We compare our results against multiple clustering baselines and find that it outperforms all of them, hence demonstrating the utility of similarity-based automated model selection for solving clustering applications.
Tabular datasets play a crucial role in various applications. Thus, developing efficient, effective, and widely compatible prediction algorithms for tabular data is important. Currently, two prominent model types, Gradient Boosted Decision Trees (GBDTs) and Deep Neural Networks (DNNs), have demonstrated performance advantages on distinct tabular prediction tasks. However, selecting an effective model for a specific tabular dataset is challenging, often demanding time-consuming hyperparameter tuning. To address this model selection dilemma, this paper proposes a new framework that amalgamates the advantages of both GBDTs and DNNs, resulting in a DNN algorithm that is as efficient as GBDTs and is competitively effective regardless of dataset preferences for GBDTs or DNNs. Our idea is rooted in an observation that deep learning (DL) offers a larger parameter space that can represent a well-performing GBDT model, yet the current back-propagation optimizer struggles to efficiently discover such optimal functionality. On the other hand, during GBDT development, hard tree pruning, entropy-driven feature gate, and model ensemble have proved to be more adaptable to tabular data. By combining these key components, we present a Tree-hybrid simple MLP (T-MLP). In our framework, a tensorized, rapidly trained GBDT feature gate, a DNN architecture pruning approach, as well as a vanilla back-propagation optimizer collaboratively train a randomly initialized MLP model. Comprehensive experiments show that T-MLP is competitive with extensively tuned DNNs and GBDTs in their dominating tabular benchmarks (88 datasets) respectively, all achieved with compact model storage and significantly reduced training duration.
Graph mining has become crucial in fields such as social science, finance, and cybersecurity. Many large-scale real-world networks exhibit both heterogeneity, where multiple node and edge types exist in the graph, and heterophily, where connected nodes may have dissimilar labels and attributes. However, existing benchmarks primarily focus on either heterophilic homogeneous graphs or homophilic heterogeneous graphs, leaving a significant gap in understanding how models perform on graphs with both heterogeneity and heterophily. To bridge this gap, we introduce H2GB, a large-scale node-classification graph benchmark that brings together the complexities of both the heterophily and heterogeneity properties of real-world graphs. H2GB encompasses 9 real-world datasets spanning 5 diverse domains, 28 baseline models, and a unified benchmarking library with a standardized data loader, evaluator, unified modeling framework, and an extensible framework for reproducibility. We establish a standardized workflow supporting both model selection and development, enabling researchers to easily benchmark graph learning methods. Extensive experiments across 28 baselines reveal that current methods struggle with heterophilic and heterogeneous graphs, underscoring the need for improved approaches. Finally, we present a new variant of the model, H2G-former, developed following our standardized workflow, that excels at this challenging benchmark. Both the benchmark and the framework are publicly available at Github and PyPI, with documentation hosted at https://junhongmit.github.io/H2GB.
Evaluation is critical for assessing capabilities, tracking scientific progress, and informing model selection. In this paper, we present three desiderata for a good benchmark for language models: (i) salience (e.g., knowledge about World War II is more salient than a random day in history), (ii) novelty (i.e., the benchmark reveals new trends in model rankings not shown by previous benchmarks), and (iii) difficulty (i.e., the benchmark should be difficult for existing models, leaving headroom for future improvement). We operationalize these three desiderata and cast benchmark creation as a search problem, that of finding benchmarks that that satisfy all three desiderata. To tackle this search problem, we present AutoBencher, which uses a language model to automatically search for datasets that meet the three desiderata. AutoBencher uses privileged information (e.g. relevant documents) to construct reliable datasets, and adaptivity with reranking to optimize for the search objective. We use AutoBencher to create datasets for math, multilingual, and knowledge-intensive question answering. The scalability of AutoBencher allows it to test fine-grained categories and tail knowledge, creating datasets that are on average 27% more novel and 22% more difficult than existing benchmarks. A closer investigation of our constructed datasets shows that we can identify specific gaps in LM knowledge in language models that are not captured by existing benchmarks, such as Gemini Pro performing much worse on question answering about the Permian Extinction and Fordism, while OpenAGI-7B performing surprisingly well on QA about COVID-19.
The choice of embedding model is a crucial step in the design of Retrieval Augmented Generation (RAG) systems. Given the sheer volume of available options, identifying clusters of similar models streamlines this model selection process. Relying solely on benchmark performance scores only allows for a weak assessment of model similarity. Thus, in this study, we evaluate the similarity of embedding models within the context of RAG systems. Our assessment is two-fold: We use Centered Kernel Alignment to compare embeddings on a pair-wise level. Additionally, as it is especially pertinent to RAG systems, we evaluate the similarity of retrieval results between these models using Jaccard and rank similarity. We compare different families of embedding models, including proprietary ones, across five datasets from the popular Benchmark Information Retrieval (BEIR). Through our experiments we identify clusters of models corresponding to model families, but interestingly, also some inter-family clusters. Furthermore, our analysis of top-k retrieval similarity reveals high-variance at low k values. We also identify possible open-source alternatives to proprietary models, with Mistral exhibiting the highest similarity to OpenAI models.
In this paper, we consider contamination by code generation test sets, in particular in their use in modern large language models. We discuss three possible sources of such contamination and show findings supporting each of them: (i) direct data leakage, (ii) indirect data leakage through the use of synthetic data and (iii) overfitting to evaluation sets during model selection. To address this, we release Less Basic Python Problems (LBPP): an uncontaminated new benchmark of 161 prompts with their associated Python solutions. LBPP is released at https://huggingface.co/datasets/CohereForAI/lbpp .
The control and modeling of robot dynamics have increasingly adopted model-free control strategies using machine learning. Given the non-linear elastic nature of bionic robotic systems, learning-based methods provide reliable alternatives by utilizing numerical data to establish a direct mapping from actuation inputs to robot trajectories without complex kinematics models. However, for developers, the method of identifying an appropriate learning model for their specific bionic robots and further constructing the transfer function has not been thoroughly discussed. Thus, this research introduces a comprehensive evaluation strategy and framework for the application of model-free control, including data collection, learning model selection, comparative analysis, and transfer function identification to effectively deal with the multi-input multi-output (MIMO) robotic data.
Tabular data is prevalent in real-world machine learning applications, and new models for supervised learning of tabular data are frequently proposed. Comparative studies assessing the performance of models typically consist of model-centric evaluation setups with overly standardized data preprocessing. This paper demonstrates that such model-centric evaluations are biased, as real-world modeling pipelines often require dataset-specific preprocessing and feature engineering. Therefore, we propose a data-centric evaluation framework. We select 10 relevant datasets from Kaggle competitions and implement expert-level preprocessing pipelines for each dataset. We conduct experiments with different preprocessing pipelines and hyperparameter optimization (HPO) regimes to quantify the impact of model selection, HPO, feature engineering, and test-time adaptation. Our main findings are: 1. After dataset-specific feature engineering, model rankings change considerably, performance differences decrease, and the importance of model selection reduces. 2. Recent models, despite their measurable progress, still significantly benefit from manual feature engineering. This holds true for both tree-based models and neural networks. 3. While tabular data is typically considered static, samples are often collected over time, and adapting to distribution shifts can be important even in supposedly static data. These insights suggest that research efforts should be directed toward a data-centric perspective, acknowledging that tabular data requires feature engineering and often exhibits temporal characteristics. Our framework is available under: https://github.com/atschalz/dc_tabeval.
Diabetes mellitus is a chronic metabolic disorder that has emerged as one of the major health problems worldwide due to its high prevalence and serious complications, which are pricey to manage. Effective management requires good glycemic control and regular follow-up in the clinic; however, non-adherence to scheduled follow-ups is very common. This study uses the Diabetes 130-US Hospitals dataset for analysis and prediction of readmission patients by various traditional machine learning models, such as XGBoost, LightGBM, CatBoost, Decision Tree, and Random Forest, and also uses an in-house LSTM neural network for comparison. The quality of the data was assured by preprocessing it, and the performance evaluation for all these models was based on accuracy, precision, recall, and F1-score. LightGBM turned out to be the best traditional model, while XGBoost was the runner-up. The LSTM model suffered from overfitting despite high training accuracy. A major strength of LSTM is capturing temporal dependencies among the patient data. Further, SHAP values were used, which improved model interpretability, whereby key factors among them number of lab procedures and discharge disposition were identified as critical in the prediction of readmissions. This study demonstrates that model selection, validation, and interpretability are key steps in predictive healthcare modeling. This will help health providers design interventions for improved follow-up adherence and better management of diabetes.
Advances in machine learning research drive progress in real-world applications. To ensure this progress, it is important to understand the potential pitfalls on the way from a novel method's success on academic benchmarks to its practical deployment. In this work, we analyze existing tabular benchmarks and find two common characteristics of tabular data in typical industrial applications that are underrepresented in the datasets usually used for evaluation in the literature. First, in real-world deployment scenarios, distribution of data often changes over time. To account for this distribution drift, time-based train/test splits should be used in evaluation. However, popular tabular datasets often lack timestamp metadata to enable such evaluation. Second, a considerable portion of datasets in production settings stem from extensive data acquisition and feature engineering pipelines. This can have an impact on the absolute and relative number of predictive, uninformative, and correlated features compared to academic datasets. In this work, we aim to understand how recent research advances in tabular deep learning transfer to these underrepresented conditions. To this end, we introduce TabReD -- a collection of eight industry-grade tabular datasets. We reassess a large number of tabular ML models and techniques on TabReD. We demonstrate that evaluation on time-based data splits leads to different methods ranking, compared to evaluation on random splits, which are common in current benchmarks. Furthermore, simple MLP-like architectures and GBDT show the best results on the TabReD datasets, while other methods are less effective in the new setting.
Filtering and annotating textual data are routine tasks in many areas, like social media or news analytics. Automating these tasks allows to scale the analyses wrt. speed and breadth of content covered and decreases the manual effort required. Due to technical advancements in Natural Language Processing, specifically the success of large foundation models, a new tool for automating such annotation processes by using a text-to-text interface given written guidelines without providing training samples has become available. In this work, we assess these advancements in-the-wild by empirically testing them in an annotation task on German Twitter data about social and political European crises. We compare the prompt-based results with our human annotation and preceding classification approaches, including Naive Bayes and a BERT-based fine-tuning/domain adaptation pipeline. Our results show that the prompt-based approach - despite being limited by local computation resources during the model selection - is comparable with the fine-tuned BERT but without any annotated training data. Our findings emphasize the ongoing paradigm shift in the NLP landscape, i.e., the unification of downstream tasks and elimination of the need for pre-labeled training data.
Since the creation of the Web, recommender systems (RSs) have been an indispensable mechanism in information filtering. State-of-the-art RSs primarily depend on categorical features, which ecoded by embedding vectors, resulting in excessively large embedding tables. To prevent over-parameterized embedding tables from harming scalability, both academia and industry have seen increasing efforts in compressing RS embeddings. However, despite the prosperity of lightweight embedding-based RSs (LERSs), a wide diversity is seen in evaluation protocols, resulting in obstacles when relating LERS performance to real-world usability. Moreover, despite the common goal of lightweight embeddings, LERSs are evaluated with a single choice between the two main recommendation tasks -- collaborative filtering and content-based recommendation. This lack of discussions on cross-task transferability hinders the development of unified, more scalable solutions. Motivated by these issues, this study investigates various LERSs' performance, efficiency, and cross-task transferability via a thorough benchmarking process. Additionally, we propose an efficient embedding compression method using magnitude pruning, which is an easy-to-deploy yet highly competitive baseline that outperforms various complex LERSs. Our study reveals the distinct performance of LERSs across the two tasks, shedding light on their effectiveness and generalizability. To support edge-based recommendations, we tested all LERSs on a Raspberry Pi 4, where the efficiency bottleneck is exposed. Finally, we conclude this paper with critical summaries of LERS performance, model selection suggestions, and underexplored challenges around LERSs for future research. To encourage future research, we publish source codes and artifacts at \href{this link}{https://github.com/chenxing1999/recsys-benchmark}.
One of the hallmark achievements of the theory of graphical models and Bayesian model selection is the celebrated greedy equivalence search (GES) algorithm due to Chickering and Meek. GES is known to consistently estimate the structure of directed acyclic graph (DAG) models in various special cases including Gaussian and discrete models, which are in particular curved exponential families. A general theory that covers general nonparametric DAG models, however, is missing. Here, we establish the consistency of greedy equivalence search for general families of DAG models that satisfy smoothness conditions on the Markov factorization, and hence may not be curved exponential families, or even parametric. The proof leverages recent advances in nonparametric Bayes to construct a test for comparing misspecified DAG models that avoids arguments based on the Laplace approximation. Nonetheless, when the Laplace approximation is valid and a consistent scoring function exists, we recover the classical result. As a result, we obtain a general consistency theorem for GES applied to general DAG models.
Semantic Communication can transform the way we transmit information, prioritizing meaningful and effective content over individual symbols or bits. This evolution promises significant benefits, including reduced latency, lower bandwidth usage, and higher throughput compared to traditional communication. However, the development of Semantic Communication faces a crucial challenge: the need for universal metrics to benchmark the joint effects of semantic information loss and energy consumption. This research introduces an innovative solution: the ``Energy-Optimized Semantic Loss'' (EOSL) function, a novel multi-objective loss function that effectively balances semantic information loss and energy consumption. Through comprehensive experiments on transformer models, including energy benchmarking, we demonstrate the remarkable effectiveness of EOSL-based model selection. We have established that EOSL-based transformer model selection achieves up to 83\% better similarity-to-power ratio (SPR) compared to BLEU score-based selection and 67\% better SPR compared to solely lowest power usage-based selection. Furthermore, we extend the applicability of EOSL to diverse and varying contexts, inspired by the principles of Meta-Learning. By cumulatively applying EOSL, we enable the model selection system to adapt to this change, leveraging historical EOSL values to guide the learning process. This work lays the foundation for energy-efficient model selection and the development of green semantic communication.
In modern wireless network architectures, such as O-RAN, artificial intelligence (AI)-based applications are deployed at intelligent controllers to carry out functionalities like scheduling or power control. The AI "apps" are selected on the basis of contextual information such as network conditions, topology, traffic statistics, and design goals. The mapping between context and AI model parameters is ideally done in a zero-shot fashion via an automatic model selection (AMS) mapping that leverages only contextual information without requiring any current data. This paper introduces a general methodology for the online optimization of AMS mappings. Optimizing an AMS mapping is challenging, as it requires exposure to data collected from many different contexts. Therefore, if carried out online, this initial optimization phase would be extremely time consuming. A possible solution is to leverage a digital twin of the physical system to generate synthetic data from multiple simulated contexts. However, given that the simulator at the digital twin is imperfect, a direct use of simulated data for the optimization of the AMS mapping would yield poor performance when tested in the real system. This paper proposes a novel method for the online optimization of AMS mapping that corrects for the bias of the simulator by means of limited real data collected from the physical system. Experimental results for a graph neural network-based power control app demonstrate the significant advantages of the proposed approach.
Predictive models may generate biased predictions when classifying imbalanced datasets. This happens when the model favors the majority class, leading to low performance in accurately predicting the minority class. To address this issue, balancing or resampling methods are critical data-centric AI approaches in the modeling process to improve prediction performance. However, there have been debates and questions about the functionality of these methods in recent years. In particular, many candidate models may exhibit very similar predictive performance, called the Rashomon effect, in model selection, and they may even produce different predictions for the same observations. Selecting one of these models without considering the predictive multiplicity -- which is the case of yielding conflicting models' predictions for any sample -- can result in blind selection. In this paper, the impact of balancing methods on predictive multiplicity is examined using the Rashomon effect. It is crucial because the blind model selection in data-centric AI is risky from a set of approximately equally accurate models. This may lead to severe problems in model selection, validation, and explanation. To tackle this matter, we conducted real dataset experiments to observe the impact of balancing methods on predictive multiplicity through the Rashomon effect by using a newly proposed metric obscurity in addition to the existing ones: ambiguity and discrepancy. Our findings showed that balancing methods inflate the predictive multiplicity and yield varying results. To monitor the trade-off between the prediction performance and predictive multiplicity for conducting the modeling process responsibly, we proposed using the extended version of the performance-gain plot when balancing the training data.
Limited accessibility to neurological care leads to underdiagnosed Parkinson's Disease (PD), preventing early intervention. Existing AI-based PD detection methods primarily focus on unimodal analysis of motor or speech tasks, overlooking the multifaceted nature of the disease. To address this, we introduce a large-scale, multi-task video dataset consisting of 1102 sessions (each containing videos of finger tapping, facial expression, and speech tasks captured via webcam) from 845 participants (272 with PD). We propose a novel Uncertainty-calibrated Fusion Network (UFNet) that leverages this multimodal data to enhance diagnostic accuracy. UFNet employs independent task-specific networks, trained with Monte Carlo Dropout for uncertainty quantification, followed by self-attended fusion of features, with attention weights dynamically adjusted based on task-specific uncertainties. To ensure patient-centered evaluation, the participants were randomly split into three sets: 60% for training, 20% for model selection, and 20% for final performance evaluation. UFNet significantly outperformed single-task models in terms of accuracy, area under the ROC curve (AUROC), and sensitivity while maintaining non-inferior specificity. Withholding uncertain predictions further boosted the performance, achieving 88.0+-0.3%$ accuracy, 93.0+-0.2% AUROC, 79.3+-0.9% sensitivity, and 92.6+-0.3% specificity, at the expense of not being able to predict for 2.3+-0.3% data (+- denotes 95% confidence interval). Further analysis suggests that the trained model does not exhibit any detectable bias across sex and ethnic subgroups and is most effective for individuals aged between 50 and 80. Requiring only a webcam and microphone, our approach facilitates accessible home-based PD screening, especially in regions with limited healthcare resources.
Despite rapid progress in large language models (LLMs), their performance on a vast majority of languages remains unsatisfactory. In this paper, we study building language-specific LLMs by adapting monolingual and multilingual LLMs. We conduct systematic experiments on how design choices (base model selection, vocabulary extension, and continued pretraining) impact the adapted LLM, both in terms of efficiency (how many tokens are needed to encode the same amount of information) and end task performance. We find that (1) the initial performance of LLM does not always correlate with the final performance after the adaptation. Adapting an English-centric models can yield better results than adapting multilingual models despite their worse initial performance on low-resource languages. (2) Efficiency can easily improved with simple vocabulary extension and continued pretraining in most LLMs we study, and (3) The optimal adaptation method (choice of the base model, new vocabulary size, training data, initialization strategy) is highly language-dependent, and the simplest embedding initialization works well across various experimental settings. Together, our work lays foundations on efficiently building language-specific LLMs by adapting existing LLMs.
aeon is a unified Python 3 library for all machine learning tasks involving time series. The package contains modules for time series forecasting, classification, extrinsic regression and clustering, as well as a variety of utilities, transformations and distance measures designed for time series data. aeon also has a number of experimental modules for tasks such as anomaly detection, similarity search and segmentation. aeon follows the scikit-learn API as much as possible to help new users and enable easy integration of aeon estimators with useful tools such as model selection and pipelines. It provides a broad library of time series algorithms, including efficient implementations of the very latest advances in research. Using a system of optional dependencies, aeon integrates a wide variety of packages into a single interface while keeping the core framework with minimal dependencies. The package is distributed under the 3-Clause BSD license and is available at https://github.com/ aeon-toolkit/aeon. This version was submitted to the JMLR journal on 02 Nov 2023 for v0.5.0 of aeon. At the time of this preprint aeon has released v0.9.0, and has had substantial changes.
This paper explores the performance of encoder and decoder language models on multilingual Natural Language Understanding (NLU) tasks, with a broad focus on Germanic languages. Building upon the ScandEval benchmark, initially restricted to evaluating encoder models, we extend the evaluation framework to include decoder models. We introduce a method for evaluating decoder models on NLU tasks and apply it to the languages Danish, Swedish, Norwegian, Icelandic, Faroese, German, Dutch, and English. Through a series of experiments and analyses, we also address research questions regarding the comparative performance of encoder and decoder models, the impact of NLU task types, and the variation across language resources. Our findings reveal that encoder models can achieve significantly better NLU performance than decoder models despite having orders of magnitude fewer parameters. Additionally, we investigate the correlation between decoders and task performance via a UMAP analysis, shedding light on the unique capabilities of decoder and encoder models. This study contributes to a deeper understanding of language model paradigms in NLU tasks and provides valuable insights for model selection and evaluation in multilingual settings.