Machine Learning最新文献

Classification with costly features (CwCF) is a classification problem that includes the cost of features in the optimization criteria. Individually for each sample, its features are sequentially acquired to maximize accuracy while minimizing the acquired features’ cost. However, existing approaches can only process data that can be expressed as vectors of fixed length. In real life, the data often possesses rich and complex structure, which can be more precisely described with formats such as XML or JSON. The data is hierarchical and often contains nested lists of objects. In this work, we extend an existing deep reinforcement learning-based algorithm with hierarchical deep sets and hierarchical softmax, so that it can directly process this data. The extended method has greater control over which features it can acquire and, in experiments with seven datasets, we show that this leads to superior performance. To showcase the real usage of the new method, we apply it to a real-life problem of classifying malicious web domains, using an online service.

Unsupervised learning methods are well established in the area of anomaly detection and achieve state of the art performances on outlier datasets. Outliers play a significant role, since they bear the potential to distort the predictions of a machine learning algorithm on a given dataset. Especially among PCA-based methods, outliers have an additional destructive potential regarding the result: they may not only distort the orientation and translation of the principal components, they also make it more complicated to detect outliers. To address this problem, we propose the robust outlier detection algorithm CoMadOut, which satisfies two required properties: (1) being robust towards outliers and (2) detecting them. Our CoMadOut outlier detection variants using comedian PCA define, dependent on its variant, an inlier region with a robust noise margin by measures of in-distribution (variant CMO) and optimized scores by measures of out-of-distribution (variants CMO*), e.g. kurtosis-weighting by CMO+k. These measures allow distribution based outlier scoring for each principal component, and thus, an appropriate alignment of the degree of outlierness between normal and abnormal instances. Experiments comparing CoMadOut with traditional, deep and other comparable robust outlier detection methods showed that the performance of the introduced CoMadOut approach is competitive to well established methods related to average precision (AP), area under the precision recall curve (AUPRC) and area under the receiver operating characteristic (AUROC) curve. In summary our approach can be seen as a robust alternative for outlier detection tasks.

Tensor decomposition has recently been gaining attention in the machine learning community for the analysis of individual traces, such as Electronic Health Records. However, this task becomes significantly more difficult when the data follows complex temporal patterns. This paper introduces the notion of a temporal phenotype as an arrangement of features over time and it proposes SWoTTeD (Sliding Window for Temporal Tensor Decomposition), a novel method to discover hidden temporal patterns. SWoTTeD integrates several constraints and regularizations to enhance the interpretability of the extracted phenotypes. We validate our proposal using both synthetic and real-world datasets, and we present an original usecase using data from the Greater Paris University Hospital. The results show that SWoTTeD achieves at least as accurate reconstruction as recent state-of-the-art tensor decomposition models, and extracts temporal phenotypes that are meaningful for clinicians.

Greedy-GQ with linear function approximation, originally proposed in Maei et al. (in: Proceedings of the international conference on machine learning (ICML), 2010), is a value-based off-policy algorithm for optimal control in reinforcement learning, and it has a non-linear two timescale structure with non-convex objective function. This paper develops its tightest finite-time error bounds. We show that the Greedy-GQ algorithm converges as fast as (mathcal {O}({1}/{sqrt{T}})) under the i.i.d. setting and (mathcal {O}({log T}/{sqrt{T}})) under the Markovian setting. We further design variant of the vanilla Greedy-GQ algorithm using the nested-loop approach, and show that its sample complexity is (mathcal {O}({log (1/epsilon )epsilon ^{-2}})), which matches with the one of the vanilla Greedy-GQ. Our finite-time error bounds match with the one of the stochastic gradient descent algorithm for general smooth non-convex optimization problems, despite of its additonal challenge in the two time-scale updates. Our finite-sample analysis provides theoretical guidance on choosing step-sizes for faster convergence in practice, and suggests the trade-off between the convergence rate and the quality of the obtained policy. Our techniques provide a general approach for finite-sample analysis of non-convex two timescale value-based reinforcement learning algorithms.

Node classification is a crucial task for efficiently analyzing graph-structured data. Related semi-supervised methods have been extensively studied to address the scarcity of labeled data in emerging classes. However, two fundamental weaknesses hinder the performance: lacking the ability to mine latent semantic information between nodes, or ignoring to simultaneously capture local and global coupling dependencies between different nodes. To solve these limitations, we propose a novel semantic-enhanced graph neural networks with global context representation for semi-supervised node classification. Specifically, we first use graph convolution network to learn short-range local dependencies, which not only considers the spatial topological structure relationship between nodes, but also takes into account the semantic correlation between nodes to enhance the representation ability of nodes. Second, an improved Transformer model is introduced to reasoning the long-range global pairwise relationships, which has linear computational complexity and is particularly important for large datasets. Finally, the proposed model shows strong performance on various open datasets, demonstrating the superiority of our solutions.

Machine learning models often struggle to generalize accurately when tested on new class distributions that were not present in their training data. This is a significant challenge for real-world applications that require quick adaptation without the need for retraining. To address this issue, few-shot learning frameworks, which includes models such as Siamese Networks, have been proposed. Siamese Networks learn similarity between pairs of records through a metric that can be easily extended to new, unseen classes. However, these systems lack interpretability, which can hinder their use in certain applications. To address this, we propose a data-agnostic method to explain the outcomes of Siamese Networks in the context of few-shot learning. Our explanation method is based on a post-hoc perturbation-based procedure that evaluates the contribution of individual input features to the final outcome. As such, it falls under the category of post-hoc explanation methods. We present two variants, one that considers each input feature independently, and another that evaluates the interplay between features. Additionally, we propose two perturbation procedures to evaluate feature contributions. Qualitative and quantitative results demonstrate that our method is able to identify highly discriminant intra-class and inter-class characteristics, as well as predictive behaviors that lead to misclassification by relying on incorrect features.

Recent studies on adversarial examples expose vulnerabilities of natural language processing models. Existing techniques for generating adversarial examples are typically driven by deterministic hierarchical rules that are agnostic to the optimal adversarial examples, a strategy that often results in adversarial samples with a suboptimal balance between magnitudes of changes and attack successes. To this end, in this research we propose two algorithms, Reversible Jump Attack (RJA) and Metropolis–Hasting Modification Reduction (MMR), to generate highly effective adversarial examples and to improve the imperceptibility of the examples, respectively. RJA utilizes a novel randomization mechanism to enlarge the search space and efficiently adapts to a number of perturbed words for adversarial examples. With these generated adversarial examples, MMR applies the Metropolis–Hasting sampler to enhance the imperceptibility of adversarial examples. Extensive experiments demonstrate that RJA-MMR outperforms current state-of-the-art methods in attack performance, imperceptibility, fluency and grammar correctness.

We devise coresets for kernel (k)-Means with a general kernel, and use them to obtain new, more efficient, algorithms. Kernel (k)-Means has superior clustering capability compared to classical (k)-Means, particularly when clusters are non-linearly separable, but it also introduces significant computational challenges. We address this computational issue by constructing a coreset, which is a reduced dataset that accurately preserves the clustering costs. Our main result is a coreset for kernel (k)-Means that works for a general kernel and has size ({{,textrm{poly},}}(kepsilon ^{-1})). Our new coreset both generalizes and greatly improves all previous results; moreover, it can be constructed in time near-linear in n. This result immediately implies new algorithms for kernel (k)-Means, such as a ((1+epsilon ))-approximation in time near-linear in n, and a streaming algorithm using space and update time ({{,textrm{poly},}}(k epsilon ^{-1} log n)). We validate our coreset on various datasets with different kernels. Our coreset performs consistently well, achieving small errors while using very few points. We show that our coresets can speed up kernel (textsc {k-Means++}) (the kernelized version of the widely used (textsc {k-Means++}) algorithm), and we further use this faster kernel (textsc {k-Means++}) for spectral clustering. In both applications, we achieve significant speedup and a better asymptotic growth while the error is comparable to baselines that do not use coresets.

Continual learning (CL) is one of the most promising trends in recent machine learning research. Its goal is to go beyond classical assumptions in machine learning and develop models and learning strategies that present high robustness in dynamic environments. This goal is realized by designing strategies that simultaneously foster the incorporation of new knowledge while avoiding forgetting past knowledge. The landscape of CL research is fragmented into several learning evaluation protocols, comprising different learning tasks, datasets, and evaluation metrics. Additionally, the benchmarks adopted so far are still distant from the complexity of real-world scenarios, and are usually tailored to highlight capabilities specific to certain strategies. In such a landscape, it is hard to clearly and objectively assess models and strategies. In this work, we fill this gap for CL on image data by introducing two novel CL benchmarks that involve multiple heterogeneous tasks from six image datasets, with varying levels of complexity and quality. Our aim is to fairly evaluate current state-of-the-art CL strategies on a common ground that is closer to complex real-world scenarios. We additionally structure our benchmarks so that tasks are presented in increasing and decreasing order of complexity—according to a curriculum—in order to evaluate if current CL models are able to exploit structure across tasks. We devote particular emphasis to providing the CL community with a rigorous and reproducible evaluation protocol for measuring the ability of a model to generalize and not to forget while learning. Furthermore, we provide an extensive experimental evaluation showing that popular CL strategies, when challenged with our proposed benchmarks, yield sub-par performance, high levels of forgetting, and present a limited ability to effectively leverage curriculum task ordering. We believe that these results highlight the need for rigorous comparisons in future CL works as well as pave the way to design new CL strategies that are able to deal with more complex scenarios.

Although deep reinforcement learning has become a promising machine learning approach for sequential decision-making problems, it is still not mature enough for high-stake domains such as autonomous driving or medical applications. In such contexts, a learned policy needs for instance to be interpretable, so that it can be inspected before any deployment (e.g., for safety and verifiability reasons). This survey provides an overview of various approaches to achieve higher interpretability in reinforcement learning (RL). To that aim, we distinguish interpretability (as an intrinsic property of a model) and explainability (as a post-hoc operation) and discuss them in the context of RL with an emphasis on the former notion. In particular, we argue that interpretable RL may embrace different facets: interpretable inputs, interpretable (transition/reward) models, and interpretable decision-making. Based on this scheme, we summarize and analyze recent work related to interpretable RL with an emphasis on papers published in the past 10 years. We also discuss briefly some related research areas and point to some potential promising research directions, notably related to the recent development of foundation models (e.g., large language models, RL from human feedback).