With the incorporation of new data gathering methods in clinical research, it becomes fundamental for survival analysis techniques to deal with high-dimensional or/and non-standard covariates. In this paper we introduce a general non-parametric independence test between right-censored survival times and covariates taking values on a general (not necessarily Euclidean) space $\mathcal{X}$. We show that our test statistic has a dual interpretation, first in terms of the supremum of a potentially infinite collection of weight-indexed log-rank tests, with weight functions belonging to a reproducing kernel Hilbert space (RKHS) of functions; and second, as the norm of the difference of embeddings of certain finite measures into the RKHS, similar to the Hilbert-Schmidt Independence Criterion (HSIC) test-statistic. We study the asymptotic properties of the test, finding sufficient conditions to ensure that our test is omnibus. The test statistic can be computed straightforwardly, and the rejection threshold is obtained via an asymptotically consistent Wild-Bootstrap procedure. We perform extensive simulations demonstrating that our testing procedure generally performs better than competing approaches in detecting complex nonlinear dependence.

Tax evasion is the illegal non-payment of taxes by individuals, corporations, and trusts. It results in a loss of state revenue that can undermine the effectiveness of government policies. One measure of tax evasion is the so-called tax gap: the difference between the income that should be reported to the tax authorities and the amount actually reported. However, economists lack a robust method for estimating the tax gap through a bottom-up approach based on fiscal audits. This is difficult because the declared tax base is available on the whole population but the income reported to the tax authorities is generally available only on a small, non-random sample of audited units. This induces a selection bias which invalidates standard statistical methods. Here, we use machine learning based on a 2-steps Gradient Boosting model, to correct for the selection bias without requiring any strong assumption on the distribution. We use our method to estimate the Italian VAT Gap related to individual firms based on information gathered from administrative sources. Our algorithm estimates the potential VAT turnover of Italian individual firms for the fiscal year 2011 and suggests that the tax gap is about 30% of the total potential tax base. Comparisons with other methods show our technique offers a significant improvement in predictive performance.

Electronic medical record (EMR) data contains historical sequences of visits of patients, and each visit contains rich information, such as patient demographics, hospital utilisation and medical codes, including diagnosis, procedure and medication codes. Most existing EMR embedding methods capture visit-code associations by constructing input visit representations as binary vectors with a static vocabulary of medical codes. With this limited representation, they fail in encapsulating rich attribute information of visits (demographics and utilisation information) and/or codes (e.g., medical code descriptions). Furthermore, current work considers visits of the same patient as discrete-time events and ignores time gaps between them. However, the time gaps between visits depict dynamics of the patient's medical history inducing varying influences on future visits. To address these limitations, we present $\mathtt{MedGraph}$, a supervised EMR embedding method that captures two types of information: (1) the visit-code associations in an attributed bipartite graph, and (2) the temporal sequencing of visits through point processes. $\mathtt{MedGraph}$ produces Gaussian embeddings for visits and codes to model the uncertainty. We evaluate the performance of $\mathtt{MedGraph}$ through an extensive experimental study and show that $\mathtt{MedGraph}$ outperforms state-of-the-art EMR embedding methods in several medical risk prediction tasks.

Drug - drug interactions ( DDIs) account for over 30% of all adverse drug reactions ( ADRs) ca ses and often occur when co - medicate more than two drugs. More alarmingly, it stays a significant ADR - mediated morbidity every year [1], and this ramps up withdrawn - risks of a drug from the market and thu s pulls a strong disincentive to drug development [2] . Though it is ideal for detecting all negative DDIs during clinical trials, DDIs - induced - ADRs cases are often reported at clinical uses and post - marketing surveillance, which pose a severe threat to public health. A study concerning the relationship between DDIs and the mortality rate of elderly hospitalized patients concludes that over 62.77% of patients present at least one DDI, and this may amount strictl y to the death of these patients [3] . Besides, DDIs also expand the length of stay and cost of hospitalization [4] .

Once users have shared their data online, it is generally difficult for them to revoke access and ask for the data to be deleted. Machine learning (ML) exacerbates this problem because any model trained with said data may have memorized it, putting users at risk of a successful privacy attack exposing their information. Yet, having models unlearn is notoriously difficult. After a data point is removed from a training set, one often resorts to entirely retraining downstream models from scratch. We introduce SISA training, a framework that decreases the number of model parameters affected by an unlearning request and caches intermediate outputs of the training algorithm to limit the number of model updates that need to be computed to have these parameters unlearn. This framework reduces the computational overhead associated with unlearning, even in the worst-case setting where unlearning requests are made uniformly across the training set. In some cases, we may have a prior on the distribution of unlearning requests that will be issued by users. We may take this prior into account to partition and order data accordingly and further decrease overhead from unlearning. Our evaluation spans two datasets from different application domains, with corresponding motivations for unlearning. Under no distributional assumptions, we observe that SISA training improves unlearning for the Purchase dataset by 3.13x, and 1.658x for the SVHN dataset, over retraining from scratch. We also validate how knowledge of the unlearning distribution provides further improvements in retraining time by simulating a scenario where we model unlearning requests that come from users of a commercial product that is available in countries with varying sensitivity to privacy. Our work contributes to practical data governance in machine learning.

The boons of machine learning have been leveraged in the industry in the past many years. With its increasing implementation, the ML tools have also evolved with time. Today, people can easily work with machine learning owing to its easy-to-use, user-friendly tools. As the gathering of data and turning it into actionable insights has been automated enough, people with some knowledge of technology and motivation can work with ML. These tools possess the strength to handle the mundane work of collecting data, adding structure and consistency where possible, and then starting the calculation.

In my first article on signal processing using machine learning, I introduced Principal Component Analysis (PCA) and Independent Component Analysis (ICA) for dimensionality reduction. We were able to see how these methods can be used to reduce the number of features in our data. However, they are linear methods: they do not always perform well when there are nonlinear relationships within our data. This is where manifold learning comes in. A manifold is any space that is locally Euclidean.

Explainable AI (XAI) is a sub-field of AI which has been gaining ground in the recent past. And as I machine learning practitioner dealing with customers day in and day out, I can see why. I've been an analytics practitioner for more than 5 years, and I swear, the hardest part of a machine learning project is not creating the perfect model which beats all the benchmarks. It's the part where you convince the customer why and how it works. Humans always had a dichotomy when faced with the unknown.

Rank-based metrics are some of the most widely used criteria for performance evaluation of computer vision models. Despite years of effort, direct optimization for these metrics remains a challenge due to their non-differentiable and non-decomposable nature. We present an efficient, theoretically sound, and general method for differentiating rank-based metrics with mini-batch gradient descent. In addition, we address optimization instability and sparsity of the supervision signal that both arise from using rank-based metrics as optimization targets. Resulting losses based on recall and Average Precision are applied to image retrieval and object detection tasks. We obtain performance that is competitive with state-of-the-art on standard image retrieval datasets and consistently improve performance of near state-of-the-art object detectors.

State-of-the-art ordinal regression methods rely on the correctness of the labels in the data. The real-world data might be susceptible to label noise, and the existing state of the art algorithms do not take label noise into account. So far, none of the approaches for ordinal regression take care of the label noise issue. We propose two novel noise models for ordinal regression. Further, we propose a general framework for robust ordinal regression learning. The proposed method is based on unbiased estimators approach and assumes the knowledge of the noise model. We then give a deep learning implementation for two commonly used loss functions for ordinal regression. We prove that this approach gives a rank consistent model, which is needed for a good ranking rule. We verify the proposed approach empirically and show that it is indeed robust to label noise. To the best of our knowledge, this is the first approach for learning robust deep ordinal regression models in the presence of label noise.