PyTorchKR Papers

최신 논문

총 9,382편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2206.06054

Specifying and Testing $k$-Safety Properties for Machine-Learning Models

Machine-learning models are becoming increasingly prevalent in our lives, for instance assisting in image-classification or decision-making tasks. Consequently, the reliability of these models is of critical importance and has resulted in the development of numerous approaches for validating and verifying their robustness and fairness. However, beyond such specific properties, it is challenging to specify, let alone check, general functional-correctness expectations from models. In this paper, we take inspiration from specifications used in formal methods, expressing functional-correctness properties by reasoning about $k$ different executions, so-called $k$-safety properties. Considering a credit-screening model of a bank, the expected property that "if a person is denied a loan and their income decreases, they should still be denied the loan" is a 2-safety property. Here, we show the wide applicability of $k$-safety properties for machine-learning models and present the first specification language for expressing them. We also operationalize the language in a framework for automatically validating such properties using metamorphic testing. Our experiments show that our framework is effective in identifying property violations, and that detected bugs could be used to train better models.

arXiv:2206.06031공식 코드 · Papers with Code

A universal synthetic dataset for machine learning on spectroscopic data

To assist in the development of machine learning methods for automated classification of spectroscopic data, we have generated a universal synthetic dataset that can be used for model validation. This dataset contains artificial spectra designed to represent experimental measurements from techniques including X-ray diffraction, nuclear magnetic resonance, and Raman spectroscopy. The dataset generation process features customizable parameters, such as scan length and peak count, which can be adjusted to fit the problem at hand. As an initial benchmark, we simulated a dataset containing 35,000 spectra based on 500 unique classes. To automate the classification of this data, eight different machine learning architectures were evaluated. From the results, we shed light on which factors are most critical to achieve optimal performance for the classification task. The scripts used to generate synthetic spectra, as well as our benchmark dataset and evaluation routines, are made publicly available to aid in the development of improved machine learning models for spectroscopic analysis.

arXiv:2206.06022

Machine Learning Training on a Real Processing-in-Memory System

Training machine learning algorithms is a computationally intensive process, which is frequently memory-bound due to repeatedly accessing large training datasets. As a result, processor-centric systems (e.g., CPU, GPU) suffer from costly data movement between memory units and processing units, which consumes large amounts of energy and execution cycles. Memory-centric computing systems, i.e., computing systems with processing-in-memory (PIM) capabilities, can alleviate this data movement bottleneck. Our goal is to understand the potential of modern general-purpose PIM architectures to accelerate machine learning training. To do so, we (1) implement several representative classic machine learning algorithms (namely, linear regression, logistic regression, decision tree, K-means clustering) on a real-world general-purpose PIM architecture, (2) characterize them in terms of accuracy, performance and scaling, and (3) compare to their counterpart implementations on CPU and GPU. Our experimental evaluation on a memory-centric computing system with more than 2500 PIM cores shows that general-purpose PIM architectures can greatly accelerate memory-bound machine learning workloads, when the necessary operations and datatypes are natively supported by PIM hardware. To our knowledge, our work is the first one to evaluate training of machine learning algorithms on a real-world general-purpose PIM architecture.

arXiv:2206.05998

GPU-Accelerated Machine Learning in Non-Orthogonal Multiple Access

Non-orthogonal multiple access (NOMA) is an interesting technology that enables massive connectivity as required in future 5G and 6G networks. While purely linear processing already achieves good performance in NOMA systems, in certain scenarios, non-linear processing is mandatory to ensure acceptable performance. In this paper, we propose a neural network architecture that combines the advantages of both linear and non-linear processing. Its real-time detection performance is demonstrated by a highly efficient implementation on a graphics processing unit (GPU). Using real measurements in a laboratory environment, we show the superiority of our approach over conventional methods.

arXiv:2206.05985

Modeling the Machine Learning Multiverse

Amid mounting concern about the reliability and credibility of machine learning research, we present a principled framework for making robust and generalizable claims: the multiverse analysis. Our framework builds upon the multiverse analysis (Steegen et al., 2016) introduced in response to psychology's own reproducibility crisis. To efficiently explore high-dimensional and often continuous ML search spaces, we model the multiverse with a Gaussian Process surrogate and apply Bayesian experimental design. Our framework is designed to facilitate drawing robust scientific conclusions about model performance, and thus our approach focuses on exploration rather than conventional optimization. In the first of two case studies, we investigate disputed claims about the relative merit of adaptive optimizers. Second, we synthesize conflicting research on the effect of learning rate on the large batch training generalization gap. For the machine learning community, the multiverse analysis is a simple and effective technique for identifying robust claims, for increasing transparency, and a step toward improved reproducibility.

arXiv:2206.05498

A Review of Causality for Learning Algorithms in Medical Image Analysis

Medical image analysis is a vibrant research area that offers doctors and medical practitioners invaluable insight and the ability to accurately diagnose and monitor disease. Machine learning provides an additional boost for this area. However, machine learning for medical image analysis is particularly vulnerable to natural biases like domain shifts that affect algorithmic performance and robustness. In this paper we analyze machine learning for medical image analysis within the framework of Technology Readiness Levels and review how causal analysis methods can fill a gap when creating robust and adaptable medical image analysis algorithms. We review methods using causality in medical imaging AI/ML and find that causal analysis has the potential to mitigate critical problems for clinical translation but that uptake and clinical downstream research has been limited so far.

arXiv:2206.05323

Memory Classifiers: Two-stage Classification for Robustness in Machine Learning

The performance of machine learning models can significantly degrade under distribution shifts of the data. We propose a new method for classification which can improve robustness to distribution shifts, by combining expert knowledge about the ``high-level" structure of the data with standard classifiers. Specifically, we introduce two-stage classifiers called memory classifiers. First, these identify prototypical data points -- memories -- to cluster the training data. This step is based on features designed with expert guidance; for instance, for image data they can be extracted using digital image processing algorithms. Then, within each cluster, we learn local classifiers based on finer discriminating features, via standard models like deep neural networks. We establish generalization bounds for memory classifiers. We illustrate in experiments that they can improve generalization and robustness to distribution shifts on image datasets. We show improvements which push beyond standard data augmentation techniques.

arXiv:2206.06279

A Machine Learning Model for Predicting, Diagnosing, and Mitigating Health Disparities in Hospital Readmission

The management of hyperglycemia in hospitalized patients has a significant impact on both morbidity and mortality. Therefore, it is important to predict the need for diabetic patients to be hospitalized. However, using standard machine learning approaches to make these predictions may result in health disparities caused by biases in the data related to social determinants (such as race, age, and gender). These biases must be removed early in the data collection process, before they enter the system and are reinforced by model predictions, resulting in biases in the model's decisions. In this paper, we propose a machine learning pipeline capable of making predictions as well as detecting and mitigating biases in the data and model predictions. This pipeline analyses the clinical data and determines whether biases exist in the data, if so, it removes those biases before making predictions. We evaluate the performance of the proposed method on a clinical dataset using accuracy and fairness measures. The findings of the results show that when we mitigate biases early during the data ingestion, we get fairer predictions.

arXiv:2206.05235

Machine Learning Inference on Inequality of Opportunity

Equality of opportunity has emerged as an important ideal of distributive justice. Empirically, Inequality of Opportunity (IOp) is measured in two steps: first, an outcome (e.g., income) is predicted given individual circumstances; and second, an inequality index (e.g., Gini) of the predictions is computed. Machine Learning (ML) methods are tremendously useful in the first step. However, they can cause sizable biases in IOp since the bias-variance trade-off allows the bias to creep in the second step. We propose a simple debiased IOp estimator robust to such ML biases and provide the first valid inferential theory for IOp. We demonstrate improved performance in simulations and report the first unbiased measures of income IOp in Europe. Mother's education and father's occupation are the circumstances that explain the most. Plug-in estimators are very sensitive to the ML algorithm, while debiased IOp estimators are robust. These results are extended to a general U-statistics setting.

arXiv:2206.05067

Model-based machine learning of critical brain dynamics

Criticality can be exactly demonstrated in certain models of brain activity, yet it remains challenging to identify in empirical data. We trained a fully connected deep neural network to learn the phases of an excitable model unfolding on the anatomical connectome of human brain. This network was then applied to brain-wide fMRI data acquired during the descent from wakefulness to deep sleep. We report high correlation between the predicted proximity to the critical point and the exponents of cluster size distributions, indicative of subcritical dynamics. This result demonstrates that conceptual models can be leveraged to identify the dynamical regime of real neural systems.

arXiv:2206.04803

Detecting Anomalous Cryptocurrency Transactions: an AML/CFT Application of Machine Learning-based Forensics

In shaping the Internet of Money, the application of blockchain and distributed ledger technologies (DLTs) to the financial sector triggered regulatory concerns. Notably, while the user anonymity enabled in this field may safeguard privacy and data protection, the lack of identifiability hinders accountability and challenges the fight against money laundering and the financing of terrorism and proliferation (AML/CFT). As law enforcement agencies and the private sector apply forensics to track crypto transfers across ecosystems that are socio-technical in nature, this paper focuses on the growing relevance of these techniques in a domain where their deployment impacts the traits and evolution of the sphere. In particular, this work offers contextualized insights into the application of methods of machine learning and transaction graph analysis. Namely, it analyzes a real-world dataset of Bitcoin transactions represented as a directed graph network through various techniques. The modeling of blockchain transactions as a complex network suggests that the use of graph-based data analysis methods can help classify transactions and identify illicit ones. Indeed, this work shows that the neural network types known as Graph Convolutional Networks (GCN) and Graph Attention Networks (GAT) are a promising AML/CFT solution. Notably, in this scenario GCN outperform other classic approaches and GAT are applied for the first time to detect anomalies in Bitcoin. Ultimately, the paper upholds the value of public-private synergies to devise forensic strategies conscious of the spirit of explainability and data openness.

arXiv:2206.05159공식 코드 · Papers with Code

An Image Processing Pipeline for Camera Trap Time-Lapse Recordings

A new open-source image processing pipeline for analyzing camera trap time-lapse recordings is described. This pipeline includes machine learning models to assist human-in-the-loop video segmentation and animal re-identification. We present some performance results and observations on the utility of this pipeline after using it in a year-long project studying the spatial ecology and social behavior of the gopher tortoise.

arXiv:2206.04934

Response to: Significance and stability of deep learning-based identification of subtypes within major psychiatric disorders. Molecular Psychiatry (2022)

Recently, Winter and Hahn [1] commented on our work on identifying subtypes of major psychiatry disorders (MPDs) based on neurobiological features using machine learning [2]. They questioned the generalizability of our methods and the statistical significance, stability, and overfitting of the results, and proposed a pipeline for disease subtyping. We appreciate their earnest consideration of our work, however, we need to point out their misconceptions of basic machine-learning concepts and delineate some key issues involved.

arXiv:2206.04740

XAudit : A Theoretical Look at Auditing with Explanations

Responsible use of machine learning requires models to be audited for undesirable properties. While a body of work has proposed using explanations for auditing, how to do so and why has remained relatively ill-understood. This work formalizes the role of explanations in auditing and investigates if and how model explanations can help audits. Specifically, we propose explanation-based algorithms for auditing linear classifiers and decision trees for feature sensitivity. Our results illustrate that Counterfactual explanations are extremely helpful for auditing. While Anchors and decision paths may not be as beneficial in the worst-case, in the average-case they do aid a lot.

arXiv:2206.04432

Discriminative and Generative Learning for Linear Estimation of Random Signals [Lecture Notes]

Inference tasks in signal processing are often characterized by the availability of reliable statistical modeling with some missing instance-specific parameters. One conventional approach uses data to estimate these missing parameters and then infers based on the estimated model. Alternatively, data can also be leveraged to directly learn the inference mapping end-to-end. These approaches for combining partially-known statistical models and data in inference are related to the notions of generative and discriminative models used in the machine learning literature, typically considered in the context of classifiers. The goal of this lecture note is to introduce the concepts of generative and discriminative learning for inference with a partially-known statistical model. While machine learning systems often lack the interpretability of traditional signal processing methods, we focus on a simple setting where one can interpret and compare the approaches in a tractable manner that is accessible and relevant to signal processing readers. In particular, we exemplify the approaches for the task of Bayesian signal estimation in a jointly Gaussian setting with the mean-squared error (MSE) objective, i.e., a linear estimation setting.

arXiv:2206.04115

Simplifying Polylogarithms with Machine Learning

Polylogrithmic functions, such as the logarithm or dilogarithm, satisfy a number of algebraic identities. For the logarithm, all the identities follow from the product rule. For the dilogarithm and higher-weight classical polylogarithms, the identities can involve five functions or more. In many calculations relevant to particle physics, complicated combinations of polylogarithms often arise from Feynman integrals. Although the initial expressions resulting from the integration usually simplify, it is often difficult to know which identities to apply and in what order. To address this bottleneck, we explore to what extent machine learning methods can help. We consider both a reinforcement learning approach, where the identities are analogous to moves in a game, and a transformer network approach, where the problem is viewed analogously to a language-translation task. While both methods are effective, the transformer network appears more powerful and holds promise for practical use in symbolic manipulation tasks in mathematical physics.

arXiv:2206.04101

What-is and How-to for Fairness in Machine Learning: A Survey, Reflection, and Perspective

Algorithmic fairness has attracted increasing attention in the machine learning community. Various definitions are proposed in the literature, but the differences and connections among them are not clearly addressed. In this paper, we review and reflect on various fairness notions previously proposed in machine learning literature, and make an attempt to draw connections to arguments in moral and political philosophy, especially theories of justice. We also consider fairness inquiries from a dynamic perspective, and further consider the long-term impact that is induced by current prediction and decision. In light of the differences in the characterized fairness, we present a flowchart that encompasses implicit assumptions and expected outcomes of different types of fairness inquiries on the data generating process, on the predicted outcome, and on the induced impact, respectively. This paper demonstrates the importance of matching the mission (which kind of fairness one would like to enforce) and the means (which spectrum of fairness analysis is of interest, what is the appropriate analyzing scheme) to fulfill the intended purpose.

arXiv:2206.03752

Machine learning-based patient selection in an emergency department

The performance of Emergency Departments (EDs) is of great importance for any health care system, as they serve as the entry point for many patients. However, among other factors, the variability of patient acuity levels and corresponding treatment requirements of patients visiting EDs imposes significant challenges on decision makers. Balancing waiting times of patients to be first seen by a physician with the overall length of stay over all acuity levels is crucial to maintain an acceptable level of operational performance for all patients. To address those requirements when assigning idle resources to patients, several methods have been proposed in the past, including the Accumulated Priority Queuing (APQ) method. The APQ method linearly assigns priority scores to patients with respect to their time in the system and acuity level. Hence, selection decisions are based on a simple system representation that is used as an input for a selection function. This paper investigates the potential of an Machine Learning (ML) based patient selection method. It assumes that for a large set of training data, including a multitude of different system states, (near) optimal assignments can be computed by a (heuristic) optimizer, with respect to a chosen performance metric, and aims to imitate such optimal behavior when applied to new situations. Thereby, it incorporates a comprehensive state representation of the system and a complex non-linear selection function. The motivation for the proposed approach is that high quality selection decisions may depend on a variety of factors describing the current state of the ED, not limited to waiting times, which can be captured and utilized by the ML model. Results show that the proposed method significantly outperforms the APQ method for a majority of evaluated settings

arXiv:2206.03584

White-box Membership Attack Against Machine Learning Based Retinopathy Classification

The advances in machine learning (ML) have greatly improved AI-based diagnosis aid systems in medical imaging. However, being based on collecting medical data specific to individuals induces several security issues, especially in terms of privacy. Even though the owner of the images like a hospital put in place strict privacy protection provisions at the level of its information system, the model trained over his images still holds disclosure potential. The trained model may be accessible to an attacker as: 1) White-box: accessing to the model architecture and parameters; 2) Black box: where he can only query the model with his own inputs through an appropriate interface. Existing attack methods include: feature estimation attacks (FEA), membership inference attack (MIA), model memorization attack (MMA) and identification attacks (IA). In this work we focus on MIA against a model that has been trained to detect diabetic retinopathy from retinal images. Diabetic retinopathy is a condition that can cause vision loss and blindness in the people who have diabetes. MIA is the process of determining whether a data sample comes from the training data set of a trained ML model or not. From a privacy perspective in our use case where a diabetic retinopathy classification model is given to partners that have at their disposal images along with patients' identifiers, inferring the membership status of a data sample can help to state if a patient has contributed or not to the training of the model.

arXiv:2206.03493공식 코드 · Papers with Code

DeepCAVE: An Interactive Analysis Tool for Automated Machine Learning

Automated Machine Learning (AutoML) is used more than ever before to support users in determining efficient hyperparameters, neural architectures, or even full machine learning pipelines. However, users tend to mistrust the optimization process and its results due to a lack of transparency, making manual tuning still widespread. We introduce DeepCAVE, an interactive framework to analyze and monitor state-of-the-art optimization procedures for AutoML easily and ad hoc. By aiming for full and accessible transparency, DeepCAVE builds a bridge between users and AutoML and contributes to establishing trust. Our framework's modular and easy-to-extend nature provides users with automatically generated text, tables, and graphic visualizations. We show the value of DeepCAVE in an exemplary use-case of outlier detection, in which our framework makes it easy to identify problems, compare multiple runs and interpret optimization processes. The package is freely available on GitHub https://github.com/automl/DeepCAVE.