This study examines the performance of item-based k-Nearest Neighbors (ItemKNN) algorithms in the RecBole and LensKit recommender system libraries. Using four data sets (Anime, Modcloth, ML-100K, and ML-1M), we assess each library's efficiency, accuracy, and scalability, focusing primarily on normalized discounted cumulative gain (nDCG). Our results show that RecBole outperforms LensKit on two of three metrics on the ML-100K data set: it achieved an 18% higher nDCG, 14% higher precision, and 35% lower recall. To ensure a fair comparison, we adjusted LensKit's nDCG calculation to match RecBole's method. This alignment made the performance more comparable, with LensKit achieving an nDCG of 0.2540 and RecBole 0.2674. Differences in similarity matrix calculations were identified as the main cause of performance deviations. After modifying LensKit to retain only the top K similar items, both libraries showed nearly identical nDCG values across all data sets. For instance, both achieved an nDCG of 0.2586 on the ML-1M data set with the same random seed. Initially, LensKit's original implementation only surpassed RecBole in the ModCloth dataset.

This paper introduces FUNGI, Features from UNsupervised GradIents, a method to enhance the features of vision encoders by leveraging self-supervised gradients. Our method is simple: given any pretrained model, we first compute gradients from various self-supervised objectives for each input. These are projected to a lower dimension and then concatenated with the model's embedding. The resulting features are evaluated on k-nearest neighbor classification over 11 datasets from vision, 5 from natural language processing, and 2 from audio. Across backbones spanning various sizes and pretraining strategies, FUNGI features provide consistent performance improvements over the embeddings. We also show that using FUNGI features can benefit linear classification and image retrieval, and that they significantly improve the retrieval-based in-context scene understanding abilities of pretrained models, for example improving upon DINO by +17% for semantic segmentation - without any training.

In supervised classification, we use a training set of labeled observations from different competing classes to form a decision rule for classifying unlabeled test set observations as accurately as possible. Starting from Fisher (1936), Rao (1948) and Fix and Hodges (1951), several parametric as well as nonparametric classifiers have been developed for this purpose (see, e.g., Duda et al., 2007; Hastie et al., 2009). Among them, the nearest neighbor classifier (see, e.g., Cover and Hart, 1967) is perhaps the most popular one. The k-nearest neighbor classifier (k-NN) classifies an observation x to the class having the maximum number of representatives among the k nearest neighbors of x. This classifier works well if the training sample size is large compared to the dimension of the data. For a suitable choice of k (which increases with the training sample size at an appropriate rate), under some mild regularity conditions, the misclassification rate of the k-NN classifier converges to the Bayes risk (i.e., the misclassification rate of the Bayes classifier) as the training sample size grows to infinity (see, e.g.

This paper introduces Neurocache, an approach to extend the effective context size of large language models (LLMs) using an external vector cache to store its past states. Like recent vector retrieval approaches, Neurocache uses an efficient k-nearest-neighbor (kNN) algorithm to retrieve relevant past states and incorporate them into the attention process. Neurocache improves upon previous methods by (1) storing compressed states, which reduces cache size; (2) performing a single retrieval operation per token which increases inference speed; and (3) extending the retrieval window to neighboring states, which improves both language modeling and downstream task accuracy. Our experiments show the effectiveness of Neurocache both for models trained from scratch and for pre-trained models such as Llama2-Figure 1: Performance and Scalability of Neurocache 7B and Mistral-7B when enhanced with the vs. Memorizing Transformers (Wu et al., 2022) on cache mechanism. We also compare Neurocache PG-19: The graph illustrates Neurocache's consistently with text retrieval methods and show lower token perplexity and faster inference times across improvements in single-document questionanswering various cache sizes on the Project Gutenberg-19 dataset, and few-shot learning tasks.

These authors contributed equally to this work. Abstract This paper explores the efficacy of Mel Frequency Cepstral Coefficients (MFCCs) in detecting abnormal heart sounds using two classification strategies: a single classifier and an ensemble classifier approach. Heart sounds were first pre-processed to remove noise and then segmented into S1, systole, S2, and diastole intervals, with thirteen MFCCs estimated from each segment, yielding 52 MFCCs per beat. Finally, MFCCs were used for heart sound classification. For that purpose, in the single classifier strategy, the MFCCs from nine consecutive beats were averaged to classify heart sounds by a single classifier (either a support vector machine (SVM), the k nearest neighbors (kNN), or a decision tree (DT)). Conversely, the ensemble classifier strategy employed nine classifiers (either nine SVMs, nine kNN classifiers, or nine DTs) to individually assess beats as normal or abnormal, with the overall classification based on the majority vote. Both methods were tested on a publicly available phonocardiogram database. The heart sound classification accuracy was 91.95% for the SVM, 91.9% for the kNN, and 87.33% for the DT in the single classifier strategy. Also, the accuracy was 93.59% for the SVM, 91.84% for the kNN, and 92.22% for the DT in the ensemble classifier strategy. Overall, the results demonstrated that the ensemble classifier strategy improved the accuracies of the DT and the SVM by 4.89% and 1.64%, establishing MFCCs as more effective than other features, including time, time-frequency, and statistical features, evaluated in similar studies.

Approximate K Nearest Neighbor (AKNN) algorithms play a pivotal role in various AI applications, including information retrieval, computer vision, and natural language processing. Although numerous AKNN algorithms and benchmarks have been developed recently to evaluate their effectiveness, the dynamic nature of realworld data presents significant challenges that existing benchmarks fail to address. Traditional benchmarks primarily assess retrieval effectiveness in static contexts and often overlook update efficiency, which is crucial for handling continuous data ingestion. This limitation results in an incomplete assessment of an AKNN algorithm's ability to adapt to changing data patterns, thereby restricting insights into their performance in dynamic environments. To address these gaps, we introduce CANDY, a benchmark tailored for Continuous Approximate Nearest Neighbor Search with Dynamic Data Ingestion. CANDY comprehensively assesses a wide range of AKNN algorithms, integrating advanced optimizations such as machine learning-driven inference to supplant traditional heuristic scans, and improved distance computation methods to reduce computational overhead. Our extensive evaluations across diverse datasets demonstrate that simpler AKNN baselines often surpass more complex alternatives in terms of recall and latency. These findings challenge established beliefs about the necessity of algorithmic complexity for high performance. Furthermore, our results underscore existing challenges and illuminate future research opportunities.

Missing Not at Random (MNAR) and nonnormal data are challenging to handle. Traditional missing data analytical techniques such as full information maximum likelihood estimation (FIML) may fail with nonnormal data as they are built on normal distribution assumptions. Two-Stage Robust Estimation (TSRE) does manage nonnormal data, but both FIML and TSRE are less explored in longitudinal studies under MNAR conditions with nonnormal distributions. Unlike traditional statistical approaches, machine learning approaches do not require distributional assumptions about the data. More importantly, they have shown promise for MNAR data; however, their application in longitudinal studies, addressing both Missing at Random (MAR) and MNAR scenarios, is also underexplored. This study utilizes Monte Carlo simulations to assess and compare the effectiveness of six analytical techniques for missing data within the growth curve modeling framework. These techniques include traditional approaches like FIML and TSRE, machine learning approaches by single imputation (K-Nearest Neighbors and missForest), and machine learning approaches by multiple imputation (micecart and miceForest). We investigate the influence of sample size, missing data rate, missing data mechanism, and data distribution on the accuracy and efficiency of model estimation. Our findings indicate that FIML is most effective for MNAR data among the tested approaches. TSRE excels in handling MAR data, while missForest is only advantageous in limited conditions with a combination of very skewed distributions, very large sample sizes (e.g., n larger than 1000), and low missing data rates.

Understanding the decision-making process of machine learning models is crucial for ensuring trustworthy machine learning. Data Shapley, a landmark study on data valuation, advances this understanding by assessing the contribution of each datum to model accuracy. However, the resource-intensive and time-consuming nature of multiple model retraining poses challenges for applying Data Shapley to large datasets. To address this, we propose the CHG (Conduct of Hardness and Gradient) score, which approximates the utility of each data subset on model accuracy during a single model training. By deriving the closed-form expression of the Shapley value for each data point under the CHG score utility function, we reduce the computational complexity to the equivalent of a single model retraining, an exponential improvement over existing methods. Additionally, we employ CHG Shapley for real-time data selection, demonstrating its effectiveness in identifying high-value and noisy data.

Data constitute the foundational component of the data economy and its marketplaces. Efficient and fair data valuation has emerged as a topic of significant interest.\ Many approaches based on marginal contribution have shown promising results in various downstream tasks. However, they are well known to be computationally expensive as they require training a large number of utility functions, which are used to evaluate the usefulness or value of a given dataset for a specific purpose. As a result, it has been recognized as infeasible to apply these methods to a data marketplace involving large-scale datasets. Consequently, a critical issue arises: how can the re-training of the utility function be avoided? To address this issue, we propose a novel data valuation method from the perspective of optimal control, named the neural dynamic data valuation (NDDV). Our method has solid theoretical interpretations to accurately identify the data valuation via the sensitivity of the data optimal control state. In addition, we implement a data re-weighting strategy to capture the unique features of data points, ensuring fairness through the interaction between data points and the mean-field states. Notably, our method requires only training once to estimate the value of all data points, significantly improving the computational efficiency. We conduct comprehensive experiments using different datasets and tasks. The results demonstrate that the proposed NDDV method outperforms the existing state-of-the-art data valuation methods in accurately identifying data points with either high or low values and is more computationally efficient.

To achieve non-parametric NMT domain adaptation, $k$-Nearest-Neighbor Machine Translation ($k$NN-MT) constructs an external datastore to store domain-specific translation knowledge, which derives a $k$NN distribution to interpolate the prediction distribution of the NMT model via a linear interpolation coefficient $\lambda$. Despite its success, $k$NN retrieval at each timestep leads to substantial time overhead. To address this issue, dominant studies resort to $k$NN-MT with adaptive retrieval ($k$NN-MT-AR), which dynamically estimates $\lambda$ and skips $k$NN retrieval if $\lambda$ is less than a fixed threshold. Unfortunately, $k$NN-MT-AR does not yield satisfactory results. In this paper, we first conduct a preliminary study to reveal two key limitations of $k$NN-MT-AR: 1) the optimization gap leads to inaccurate estimation of $\lambda$ for determining $k$NN retrieval skipping, and 2) using a fixed threshold fails to accommodate the dynamic demands for $k$NN retrieval at different timesteps. To mitigate these limitations, we then propose $k$NN-MT with dynamic retrieval ($k$NN-MT-DR) that significantly extends vanilla $k$NN-MT in two aspects. Firstly, we equip $k$NN-MT with a MLP-based classifier for determining whether to skip $k$NN retrieval at each timestep. Particularly, we explore several carefully-designed scalar features to fully exert the potential of the classifier. Secondly, we propose a timestep-aware threshold adjustment method to dynamically generate the threshold, which further improves the efficiency of our model. Experimental results on the widely-used datasets demonstrate the effectiveness and generality of our model.\footnote{Our code is available at \url{https://github.com/DeepLearnXMU/knn-mt-dr}.