The choice of a graph learning (GL) model (i.e., a GL algorithm and its hyperparameter settings) has a significant impact on the performance of downstream tasks. However, selecting the right GL model becomes increasingly difficult and time consuming as more and more GL models are developed. Accordingly, it is of great significance and practical value to equip users of GL with the ability to perform a near-instantaneous selection of an effective GL model without manual intervention. Despite the recent attempts to tackle this important problem, there has been no comprehensive benchmark environment to evaluate the performance of GL model selection methods. To bridge this gap, we present GLEMOS in this work, a comprehensive benchmark for instantaneous GL model selection that makes the following contributions. (i) GLEMOS provides extensive benchmark data for fundamental GL tasks, i.e., link prediction and node classification, including the performances of 366 models on 457 graphs on these tasks. (ii) GLEMOS designs multiple evaluation settings, and assesses how effectively representative model selection techniques perform in these different settings. (iii) GLEMOS is designed to be easily extended with new models, new graphs, and new performance records. (iv) Based on the experimental results, we discuss the limitations of existing approaches and highlight future research directions. To promote research on this significant problem, we make the benchmark data and code publicly available at https://github.com/facebookresearch/glemos.
We propose a novel model-selection method for dynamic networks. Our approach involves training a classifier on a large body of synthetic network data. The data is generated by simulating nine state-of-the-art random graph models for dynamic networks, with parameter range chosen to ensure exponential growth of the network size in time. We design a conceptually novel type of dynamic features that count new links received by a group of vertices in a particular time interval. The proposed features are easy to compute, analytically tractable, and interpretable. Our approach achieves a near-perfect classification of synthetic networks, exceeding the state-of-the-art by a large margin. Applying our classification method to real-world citation networks gives credibility to the claims in the literature that models with preferential attachment, fitness and aging fit real-world citation networks best, although sometimes, the predicted model does not involve vertex fitness.
Conformal prediction equips machine learning models with a reasonable notion of uncertainty quantification without making strong distributional assumptions. It wraps around any black-box prediction model and converts point predictions into set predictions that have a predefined marginal coverage guarantee. However, conformal prediction only works if we fix the underlying machine learning model in advance. A relatively unaddressed issue in conformal prediction is that of model selection and/or aggregation: for a given problem, which of the plethora of prediction methods (random forests, neural nets, regularized linear models, etc.) should we conformalize? This paper proposes a new approach towards conformal model aggregation in online settings that is based on combining the prediction sets from several algorithms by voting, where weights on the models are adapted over time based on past performance.
Synthetic users are cost-effective proxies for real users in the evaluation of conversational recommender systems. Large language models show promise in simulating human-like behavior, raising the question of their ability to represent a diverse population of users. We introduce a new protocol to measure the degree to which language models can accurately emulate human behavior in conversational recommendation. This protocol is comprised of five tasks, each designed to evaluate a key property that a synthetic user should exhibit: choosing which items to talk about, expressing binary preferences, expressing open-ended preferences, requesting recommendations, and giving feedback. Through evaluation of baseline simulators, we demonstrate these tasks effectively reveal deviations of language models from human behavior, and offer insights on how to reduce the deviations with model selection and prompting strategies.
Enabling efficient and accurate deep neural network (DNN) inference on microcontrollers is non-trivial due to the constrained on-chip resources. Current methodologies primarily focus on compressing larger models yet at the expense of model accuracy. In this paper, we rethink the problem from the inverse perspective by constructing small/weak models directly and improving their accuracy. Thus, we introduce DiTMoS, a novel DNN training and inference framework with a selector-classifiers architecture, where the selector routes each input sample to the appropriate classifier for classification. DiTMoS is grounded on a key insight: a composition of weak models can exhibit high diversity and the union of them can significantly boost the accuracy upper bound. To approach the upper bound, DiTMoS introduces three strategies including diverse training data splitting to increase the classifiers' diversity, adversarial selector-classifiers training to ensure synergistic interactions thereby maximizing their complementarity, and heterogeneous feature aggregation to improve the capacity of classifiers. We further propose a network slicing technique to alleviate the extra memory overhead incurred by feature aggregation. We deploy DiTMoS on the Neucleo STM32F767ZI board and evaluate it based on three time-series datasets for human activity recognition, keywords spotting, and emotion recognition, respectively. The experiment results manifest that: (a) DiTMoS achieves up to 13.4% accuracy improvement compared to the best baseline; (b) network slicing almost completely eliminates the memory overhead incurred by feature aggregation with a marginal increase of latency.
Decision-makers are often experts of their domain and take actions based on their domain knowledge. Doctors, for instance, may prescribe treatments by predicting the likely outcome of each available treatment. Actions of an expert thus naturally encode part of their domain knowledge, and can help make inferences within the same domain: Knowing doctors try to prescribe the best treatment for their patients, we can tell treatments prescribed more frequently are likely to be more effective. Yet in machine learning, the fact that most decision-makers are experts is often overlooked, and "expertise" is seldom leveraged as an inductive bias. This is especially true for the literature on treatment effect estimation, where often the only assumption made about actions is that of overlap. In this paper, we argue that expertise - particularly the type of expertise the decision-makers of a domain are likely to have - can be informative in designing and selecting methods for treatment effect estimation. We formally define two types of expertise, predictive and prognostic, and demonstrate empirically that: (i) the prominent type of expertise in a domain significantly influences the performance of different methods in treatment effect estimation, and (ii) it is possible to predict the type of expertise present in a dataset, which can provide a quantitative basis for model selection.
The growing demand for personalized decision-making has led to a surge of interest in estimating the Conditional Average Treatment Effect (CATE). Various types of CATE estimators have been developed with advancements in machine learning and causal inference. However, selecting the desirable CATE estimator through a conventional model validation procedure remains impractical due to the absence of counterfactual outcomes in observational data. Existing approaches for CATE estimator selection, such as plug-in and pseudo-outcome metrics, face two challenges. First, they must determine the metric form and the underlying machine learning models for fitting nuisance parameters (e.g., outcome function, propensity function, and plug-in learner). Second, they lack a specific focus on selecting a robust CATE estimator. To address these challenges, this paper introduces a Distributionally Robust Metric (DRM) for CATE estimator selection. The proposed DRM is nuisance-free, eliminating the need to fit models for nuisance parameters, and it effectively prioritizes the selection of a distributionally robust CATE estimator. The experimental results validate the effectiveness of the DRM method in selecting CATE estimators that are robust to the distribution shift incurred by covariate shift and hidden confounders.
This paper presents a benchmark self-evolving framework to dynamically evaluate rapidly advancing Large Language Models (LLMs), aiming for a more accurate assessment of their capabilities and limitations. We utilize a multi-agent system to manipulate the context or question of original instances, reframing new evolving instances with high confidence that dynamically extend existing benchmarks. Towards a more scalable, robust and fine-grained evaluation, we implement six reframing operations to construct evolving instances testing LLMs against diverse queries, data noise and probing their problem-solving sub-abilities. With this framework, we extend benchmark datasets of four tasks. Experimental results show a general performance decline in most LLMs against their original results. This decline under our scalable and robust evaluations, alongside our fine-grained evaluation, more accurately reflect models' capabilities. Besides, our framework widens performance discrepancies both between different models and within the same model across various tasks, facilitating more informed model selection for specific tasks (Code and data are available at https://github.com/NanshineLoong/Self-Evolving-Benchmark).
We investigate model assessment and selection in a changing environment, by synthesizing datasets from both the current time period and historical epochs. To tackle unknown and potentially arbitrary temporal distribution shift, we develop an adaptive rolling window approach to estimate the generalization error of a given model. This strategy also facilitates the comparison between any two candidate models by estimating the difference of their generalization errors. We further integrate pairwise comparisons into a single-elimination tournament, achieving near-optimal model selection from a collection of candidates. Theoretical analyses and numerical experiments demonstrate the adaptivity of our proposed methods to the non-stationarity in data.
Large language models (LLMs) have recently received considerable attention as alternative solutions for task planning. However, comparing the performance of language-oriented task planners becomes difficult, and there exists a dearth of detailed exploration regarding the effects of various factors such as pre-trained model selection and prompt construction. To address this, we propose a benchmark system for automatically quantifying performance of task planning for home-service embodied agents. Task planners are tested on two pairs of datasets and simulators: 1) ALFRED and AI2-THOR, 2) an extension of Watch-And-Help and VirtualHome. Using the proposed benchmark system, we perform extensive experiments with LLMs and prompts, and explore several enhancements of the baseline planner. We expect that the proposed benchmark tool would accelerate the development of language-oriented task planners.
Recent studies have highlighted the benefits of generating multiple synthetic datasets for supervised learning, from increased accuracy to more effective model selection and uncertainty estimation. These benefits have clear empirical support, but the theoretical understanding of them is currently very light. We seek to increase the theoretical understanding by deriving bias-variance decompositions for several settings of using multiple synthetic datasets, including differentially private synthetic data. Our theory yields a simple rule of thumb to select the appropriate number of synthetic datasets in the case of mean-squared error and Brier score. We investigate how our theory works in practice with several real datasets, downstream predictors and error metrics. As our theory predicts, multiple synthetic datasets often improve accuracy, while a single large synthetic dataset gives at best minimal improvement, showing that our insights are practically relevant.
Tensor networks are efficient for extremely high-dimensional representation, but their model selection, known as tensor network structure search (TN-SS), is a challenging problem. Although several works have targeted TN-SS, most existing algorithms are manually crafted heuristics with poor performance, suffering from the curse of dimensionality and local convergence. In this work, we jump out of the box, studying how to harness large language models (LLMs) to automatically discover new TN-SS algorithms, replacing the involvement of human experts. By observing how human experts innovate in research, we model their common workflow and propose an automatic algorithm discovery framework called tnGPS. The proposed framework is an elaborate prompting pipeline that instruct LLMs to generate new TN-SS algorithms through iterative refinement and enhancement. The experimental results demonstrate that the algorithms discovered by tnGPS exhibit superior performance in benchmarks compared to the current state-of-the-art methods.
Large Language Model (LLM) leaderboards based on benchmark rankings are regularly used to guide practitioners in model selection. Often, the published leaderboard rankings are taken at face value - we show this is a (potentially costly) mistake. Under existing leaderboards, the relative performance of LLMs is highly sensitive to (often minute) details. We show that for popular multiple-choice question benchmarks (e.g., MMLU), minor perturbations to the benchmark, such as changing the order of choices or the method of answer selection, result in changes in rankings up to 8 positions. We explain this phenomenon by conducting systematic experiments over three broad categories of benchmark perturbations and identifying the sources of this behavior. Our analysis results in several best-practice recommendations, including the advantage of a hybrid scoring method for answer selection. Our study highlights the dangers of relying on simple benchmark evaluations and charts the path for more robust evaluation schemes on the existing benchmarks. The code for this paper is available at https://github.com/National-Center-for-AI-Saudi-Arabia/lm-evaluation-harness.
Spiking neural networks (SNNs), inspired by the spiking behavior of biological neurons, provide a unique pathway for capturing the intricacies of temporal data. However, applying SNNs to time-series forecasting is challenging due to difficulties in effective temporal alignment, complexities in encoding processes, and the absence of standardized guidelines for model selection. In this paper, we propose a framework for SNNs in time-series forecasting tasks, leveraging the efficiency of spiking neurons in processing temporal information. Through a series of experiments, we demonstrate that our proposed SNN-based approaches achieve comparable or superior results to traditional time-series forecasting methods on diverse benchmarks with much less energy consumption. Furthermore, we conduct detailed analysis experiments to assess the SNN's capacity to capture temporal dependencies within time-series data, offering valuable insights into its nuanced strengths and effectiveness in modeling the intricate dynamics of temporal data. Our study contributes to the expanding field of SNNs and offers a promising alternative for time-series forecasting tasks, presenting a pathway for the development of more biologically inspired and temporally aware forecasting models. Our code is available at https://github.com/microsoft/SeqSNN.
In recent years, there has been an increasing recognition that when machine learning (ML) algorithms are used to automate decisions, they may mistreat individuals or groups, with legal, ethical, or economic implications. Recommender systems are prominent examples of these machine learning (ML) systems that aid users in making decisions. The majority of past literature research on RS fairness treats user and item fairness concerns independently, ignoring the fact that recommender systems function in a two-sided marketplace. In this paper, we propose CP-FairRank, an optimization-based re-ranking algorithm that seamlessly integrates fairness constraints from both the consumer and producer side in a joint objective framework. The framework is generalizable and may take into account varied fairness settings based on group segmentation, recommendation model selection, and domain, which is one of its key characteristics. For instance, we demonstrate that the system may jointly increase consumer and producer fairness when (un)protected consumer groups are defined on the basis of their activity level and main-streamness, while producer groups are defined according to their popularity level. For empirical validation, through large-scale on eight datasets and four mainstream collaborative filtering (CF) recommendation models, we demonstrate that our proposed strategy is able to improve both consumer and producer fairness without compromising or very little overall recommendation quality, demonstrating the role algorithms may play in avoiding data biases.
In the realm of Earth science, effective cloud property retrieval, encompassing cloud masking, cloud phase classification, and cloud optical thickness (COT) prediction, remains pivotal. Traditional methodologies necessitate distinct models for each sensor instrument due to their unique spectral characteristics. Recent strides in Earth Science research have embraced machine learning and deep learning techniques to extract features from satellite datasets' spectral observations. However, prevailing approaches lack novel architectures accounting for hierarchical relationships among retrieval tasks. Moreover, considering the spectral diversity among existing sensors, the development of models with robust generalization capabilities over different sensor datasets is imperative. Surprisingly, there is a dearth of methodologies addressing the selection of an optimal model for diverse datasets. In response, this paper introduces MT-HCCAR, an end-to-end deep learning model employing multi-task learning to simultaneously tackle cloud masking, cloud phase retrieval (classification tasks), and COT prediction (a regression task). The MT-HCCAR integrates a hierarchical classification network (HC) and a classification-assisted attention-based regression network (CAR), enhancing precision and robustness in cloud labeling and COT prediction. Additionally, a comprehensive model selection method rooted in K-fold cross-validation, one standard error rule, and two introduced performance scores is proposed to select the optimal model over three simulated satellite datasets OCI, VIIRS, and ABI. The experiments comparing MT-HCCAR with baseline methods, the ablation studies, and the model selection affirm the superiority and the generalization capabilities of MT-HCCAR.
Machine learning (ML) models are only as good as the data they are trained on. But recent studies have found datasets widely used to train and evaluate ML models, e.g. ImageNet, to have pervasive labeling errors. Erroneous labels on the train set hurt ML models' ability to generalize, and they impact evaluation and model selection using the test set. Consequently, learning in the presence of labeling errors is an active area of research, yet this field lacks a comprehensive benchmark to evaluate these methods. Most of these methods are evaluated on a few computer vision datasets with significant variance in the experimental protocols. With such a large pool of methods and inconsistent evaluation, it is also unclear how ML practitioners can choose the right models to assess label quality in their data. To this end, we propose a benchmarking environment AQuA to rigorously evaluate methods that enable machine learning in the presence of label noise. We also introduce a design space to delineate concrete design choices of label error detection models. We hope that our proposed design space and benchmark enable practitioners to choose the right tools to improve their label quality and that our benchmark enables objective and rigorous evaluation of machine learning tools facing mislabeled data.
Various methods have recently been proposed to estimate causal effects with confidence intervals that are uniformly valid over a set of data generating processes when high-dimensional nuisance models are estimated by post-model-selection or machine learning estimators. These methods typically require that all the confounders are observed to ensure identification of the effects. We contribute by showing how valid semiparametric inference can be obtained in the presence of unobserved confounders and high-dimensional nuisance models. We propose uncertainty intervals which allow for unobserved confounding, and show that the resulting inference is valid when the amount of unobserved confounding is small relative to the sample size; the latter is formalized in terms of convergence rates. Simulation experiments illustrate the finite sample properties of the proposed intervals and investigate an alternative procedure that improves the empirical coverage of the intervals when the amount of unobserved confounding is large. Finally, a case study on the effect of smoking during pregnancy on birth weight is used to illustrate the use of the methods introduced to perform a sensitivity analysis to unobserved confounding.
Large language models (LLMs) have demonstrated impressive capabilities in various natural language processing tasks. Despite this, their application to information retrieval (IR) tasks is still challenging due to the infrequent occurrence of many IR-specific concepts in natural language. While prompt-based methods can provide task descriptions to LLMs, they often fall short in facilitating a comprehensive understanding and execution of IR tasks, thereby limiting LLMs' applicability. To address this gap, in this work, we explore the potential of instruction tuning to enhance LLMs' proficiency in IR tasks. We introduce a novel instruction tuning dataset, INTERS, encompassing 20 tasks across three fundamental IR categories: query understanding, document understanding, and query-document relationship understanding. The data are derived from 43 distinct datasets with manually written templates. Our empirical results reveal that INTERS significantly boosts the performance of various publicly available LLMs, such as LLaMA, Mistral, and Phi, in IR tasks. Furthermore, we conduct extensive experiments to analyze the effects of instruction design, template diversity, few-shot demonstrations, and the volume of instructions on performance. We make our dataset and the fine-tuned models publicly accessible at https://github.com/DaoD/INTERS.
This paper investigates an under-explored but important problem: given a collection of pre-trained neural networks, predicting their performance on each multi-modal task without fine-tuning them, such as image recognition, referring, captioning, visual question answering, and text question answering. A brute-force approach is to finetune all models on all target datasets, bringing high computational costs. Although recent-advanced approaches employed lightweight metrics to measure models' transferability,they often depend heavily on the prior knowledge of a single task, making them inapplicable in a multi-modal multi-task scenario. To tackle this issue, we propose an efficient multi-task model selector (EMMS), which employs large-scale foundation models to transform diverse label formats such as categories, texts, and bounding boxes of different downstream tasks into a unified noisy label embedding. EMMS can estimate a model's transferability through a simple weighted linear regression, which can be efficiently solved by an alternating minimization algorithm with a convergence guarantee. Extensive experiments on 5 downstream tasks with 24 datasets show that EMMS is fast, effective, and generic enough to assess the transferability of pre-trained models, making it the first model selection method in the multi-task scenario. For instance, compared with the state-of-the-art method LogME enhanced by our label embeddings, EMMS achieves 9.0\%, 26.3\%, 20.1\%, 54.8\%, 12.2\% performance gain on image recognition, referring, captioning, visual question answering, and text question answering, while bringing 5.13x, 6.29x, 3.59x, 6.19x, and 5.66x speedup in wall-clock time, respectively. The code is available at https://github.com/OpenGVLab/Multitask-Model-Selector.