Having built the Machine Learning model, we need to evaluate it.We can set targets for these metrics in the beginning of the Data Science or Machine Learning project. Achievement of these targets can be considered as one of the project success criteria. Evaluation metrics are used to measure the quality of the statistical or machine learning model. Evaluating machine learning models or algorithms is essential for any project. There are many different types of evaluation metrics available to test a model.

Facial recognition tools are becoming exceptionally accurate in identifying people from images. However, this comes at the cost of privacy for users of online services with photo management (e.g. social media platforms). Particularly troubling is the ability to leverage unsupervised learning to recognize faces even when the user has not labeled their images. In this paper we propose Ulixes, a strategy to generate visually non-invasive facial noise masks that yield adversarial examples, preventing the formation of identifiable user clusters in the embedding space of facial encoders. This is applicable even when a user is unmasked and labeled images are available online. We demonstrate the effectiveness of Ulixes by showing that various classification and clustering methods cannot reliably label the adversarial examples we generate. We also study the effects of Ulixes in various black-box settings and compare it to the current state of the art in adversarial machine learning. Finally, we challenge the effectiveness of Ulixes against adversarially trained models and show that it is robust to countermeasures.

Patient-derived extracellular vesicles (EVs) that contains a complex biological cargo is a valuable source of liquid biopsy diagnostics to aid in early detection, cancer screening, and precision nanotherapeutics. In this study, we predicted that coupling cancer patient blood-derived EVs to time-resolved spectroscopy and artificial intelligence (AI) could provide a robust cancer screening and follow-up tools. Methods: Fluorescence correlation spectroscopy (FCS) measurements were performed on 24 blood samples-derived EVs. Blood samples were obtained from 15 cancer patients (presenting 5 different types of cancers), and 9 healthy controls (including patients with benign lesions). The obtained FCS autocorrelation spectra were processed into power spectra using the Fast-Fourier Transform algorithm and subjected to various machine learning algorithms to distinguish cancer spectra from healthy control spectra. Results and Applications: The performance of AdaBoost Random Forest (RF) classifier, support vector machine, and multilayer perceptron, were tested on selected frequencies in the N=118 power spectra. The RF classifier exhibited a 90% classification accuracy and high sensitivity and specificity in distinguishing the FCS power spectra of cancer patients from those of healthy controls. Further, an image convolutional neural network (CNN), ResNet network, and a quantum CNN were assessed on the power spectral images as additional validation tools. All image-based CNNs exhibited a nearly equal classification performance with an accuracy of roughly 82% and reasonably high sensitivity and specificity scores. Our pilot study demonstrates that AI-algorithms coupled to time-resolved FCS power spectra can accurately and differentially classify the complex patient-derived EVs from different cancer samples of distinct tissue subtypes.

Photoplethysmography (PPG) signal comprises physiological information related to cardiorespiratory health. However, while recording, these PPG signals are easily corrupted by motion artifacts and body movements, leading to noise enriched, poor quality signals. Therefore ensuring high-quality signals is necessary to extract cardiorespiratory information accurately. Although there exists several rule-based and Machine-Learning (ML) - based approaches for PPG signal quality estimation, those algorithms' efficacy is questionable. Thus, this work proposes a lightweight CNN architecture for signal quality assessment employing a novel Quantum pattern recognition (QPR) technique. The proposed algorithm is validated on manually annotated data obtained from the University of Queensland database. A total of 28366, 5s signal segments are preprocessed and transformed into image files of 20 x 500 pixels. The image files are treated as an input to the 2D CNN architecture. The developed model classifies the PPG signal as `good' or `bad' with an accuracy of 98.3% with 99.3% sensitivity, 94.5% specificity and 98.9% F1-score. Finally, the performance of the proposed framework is validated against the noisy `Welltory app' collected PPG database. Even in a noisy environment, the proposed architecture proved its competence. Experimental analysis concludes that a slim architecture along with a novel Spatio-temporal pattern recognition technique improve the system's performance. Hence, the proposed approach can be useful to classify good and bad PPG signals for a resource-constrained wearable implementation.

Tree-based models such as decision trees and random forests (RF) are a cornerstone of modern machine-learning practice. To mitigate overfitting, trees are typically regularized by a variety of techniques that modify their structure (e.g. pruning). We introduce Hierarchical Shrinkage (HS), a post-hoc algorithm that does not modify the tree structure, and instead regularizes the tree by shrinking the prediction over each node towards the sample means of its ancestors. The amount of shrinkage is controlled by a single regularization parameter and the number of data points in each ancestor. Since HS is a post-hoc method, it is extremely fast, compatible with any tree growing algorithm, and can be used synergistically with other regularization techniques. Extensive experiments over a wide variety of real-world datasets show that HS substantially increases the predictive performance of decision trees, even when used in conjunction with other regularization techniques. Moreover, we find that applying HS to each tree in an RF often improves accuracy, as well as its interpretability by simplifying and stabilizing its decision boundaries and SHAP values. We further explain the success of HS in improving prediction performance by showing its equivalence to ridge regression on a (supervised) basis constructed of decision stumps associated with the internal nodes of a tree. All code and models are released in a full-fledged package available on Github (github.com/csinva/imodels)

Commercial banks receive a lot of applications for credit cards. Many of them get rejected for many reasons, like high loan balances, low-income levels, or too many inquiries on an individual's credit report, for example. Manually analyzing these applications is mundane, error-prone, and time-consuming (and time is money!). Luckily, this task can be automated with the power of machine learning, and pretty much every commercial bank does so nowadays. In this article, we will build an automatic credit card approval predictor using machine learning techniques, just like real banks do.

Device fingerprinting combined with Machine and Deep Learning (ML/DL) report promising performance when detecting cyberattacks targeting data managed by resource-constrained spectrum sensors. However, the amount of data needed to train models and the privacy concerns of such scenarios limit the applicability of centralized ML/DL-based approaches. Federated learning (FL) addresses these limitations by creating federated and privacy-preserving models. However, FL is vulnerable to malicious participants, and the impact of adversarial attacks on federated models detecting spectrum sensing data falsification (SSDF) attacks on spectrum sensors has not been studied. To address this challenge, the first contribution of this work is the creation of a novel dataset suitable for FL and modeling the behavior (usage of CPU, memory, or file system, among others) of resource-constrained spectrum sensors affected by different SSDF attacks. The second contribution is a pool of experiments analyzing and comparing the robustness of federated models according to i) three families of spectrum sensors, ii) eight SSDF attacks, iii) four scenarios dealing with unsupervised (anomaly detection) and supervised (binary classification) federated models, iv) up to 33% of malicious participants implementing data and model poisoning attacks, and v) four aggregation functions acting as anti-adversarial mechanisms to increase the models robustness.

Encoding facts as representations of entities and binary relationships between them, as learned by knowledge graph representation models, is useful for various tasks, including predicting new facts, question answering, fact checking and information retrieval. The focus of this thesis is on (i) improving knowledge graph representation with the aim of tackling the link prediction task; and (ii) devising a theory on how semantics can be captured in the geometry of relation representations. Most knowledge graphs are very incomplete and manually adding new information is costly, which drives the development of methods which can automatically infer missing facts. The first contribution of this thesis is HypER, a convolutional model which simplifies and improves upon the link prediction performance of the existing convolutional state-of-the-art model ConvE and can be mathematically explained in terms of constrained tensor factorisation. The second contribution is TuckER, a relatively straightforward linear model, which, at the time of its introduction, obtained state-of-the-art link prediction performance across standard datasets. The third contribution is MuRP, first multi-relational graph representation model embedded in hyperbolic space. MuRP outperforms all existing models and its Euclidean counterpart MuRE in link prediction on hierarchical knowledge graph relations whilst requiring far fewer dimensions. Despite the development of a large number of knowledge graph representation models with gradually increasing predictive performance, relatively little is known of the latent structure they learn. We generalise recent theoretical understanding of how semantic relations of similarity, paraphrase and analogy are encoded in the geometric interactions of word embeddings to how more general relations, as found in knowledge graphs, can be encoded in their representations.

Pseudo-labeling solutions for positive-unlabeled (PU) learning have the potential to result in higher performance compared to cost-sensitive learning but are vulnerable to incorrectly estimated pseudo-labels. In this paper, we provide a theoretical analysis of a risk estimator that combines risk on PU and pseudo-labeled data. Furthermore, we show analytically as well as experimentally that such an estimator results in lower excess risk compared to using PU data alone, provided that enough samples are pseudo-labeled with acceptable error rates. We then propose PUUPL, a novel training procedure for PU learning that leverages the epistemic uncertainty of an ensemble of deep neural networks to minimize errors in pseudo-label selection. We conclude with extensive experiments showing the effectiveness of our proposed algorithm over different datasets, modalities, and learning tasks. These show that PUUPL enables a reduction of up to 20% in test error rates even when prior and negative samples are not provided for validation, setting a new state-of-the-art for PU learning.

In domains ranging from computer vision to natural language processing, machine learning models have been shown to exhibit stark disparities, often performing worse for members of traditionally underserved groups. One factor contributing to these performance gaps is a lack of representation in the data the models are trained on. It is often unclear, however, how to operationalize representativeness in specific applications. Here we formalize the problem of creating equitable training datasets, and propose a statistical framework for addressing this problem. We consider a setting where a model builder must decide how to allocate a fixed data collection budget to gather training data from different subgroups. We then frame dataset creation as a constrained optimization problem, in which one maximizes a function of group-specific performance metrics based on (estimated) group-specific learning rates and costs per sample. This flexible approach incorporates preferences of model-builders and other stakeholders, as well as the statistical properties of the learning task. When data collection decisions are made sequentially, we show that under certain conditions this optimization problem can be efficiently solved even without prior knowledge of the learning rates. To illustrate our approach, we conduct a simulation study of polygenic risk scores on synthetic genomic data -- an application domain that often suffers from non-representative data collection. We find that our adaptive sampling strategy outperforms several common data collection heuristics, including equal and proportional sampling, demonstrating the value of strategic dataset design for building equitable models.