In biostatistics, for each of the specific situation, statistical methods are available for analysis and interpretation of the data. To select the appropriate statistical method, one need to know the assumption and conditions of the statistical methods, so that proper statistical method can be selected for data analysis. Two main statistical methods are used in data analysis: descriptive statistics, which summarizes data using indexes such as mean and median and another is inferential statistics, which draw conclusions from data using statistical tests such as student's t-test. Selection of appropriate statistical method depends on the following three things: Aim and objective of the study, Type and distribution of the data used, and Nature of the observations (paired/unpaired). All type of statistical methods that are used to compare the means are called parametric while statistical methods used to compare other than means (ex-median/mean ranks/proportions) are called nonparametric methods.

The book starts off with a look at what machine learning is and what sorts of things it can be used for. The actual subject of the book gets started at Chapter 3 where we meet linear regression. I'm not at all sure I class regression as ML - it is more classical statistics. What is interesting about this presentation is that the regression problem is solved using gradient descent, i.e an iterative algorithm like many others in machine learning. This is a good idea in that it is a simple way into the idea of iterative learning.

Multivariate Adaptive Regression Splines or commonly known as MARS is an algorithm best suited for high dimensional and complex non-linear relationship dataset. It can be seen as a generalised form of Stepwise Regression (Stepwise regression does a forward selection first where it starts loading the model and then pruning or backward selection to remove the variables that don't help reduce the error rate significantly). This function is similar to rectified linear function of neural network where the starting point is the input itself or 0. MARS uses this function to create knots and this is known as Spline. These functions are always generated in pair of Left function and Right function. MARS then generates multiple such functions that we call basis function for all the input variables and then runs linear regression on each basis function's output.

Federated learning (FL) has received a surge of interest in recent years thanks to its benefits in data privacy protection, efficient communication, and parallel data processing. Also, with appropriate algorithmic designs, one could achieve the desirable linear speedup for convergence effect in FL. However, most existing works on FL are limited to systems with i.i.d. data and centralized parameter servers and results on decentralized FL with heterogeneous datasets remains limited. Moreover, whether or not the linear speedup for convergence is achievable under fully decentralized FL with data heterogeneity remains an open question. In this paper, we address these challenges by proposing a new algorithm, called NET-FLEET, for fully decentralized FL systems with data heterogeneity. The key idea of our algorithm is to enhance the local update scheme in FL (originally intended for communication efficiency) by incorporating a recursive gradient correction technique to handle heterogeneous datasets. We show that, under appropriate parameter settings, the proposed NET-FLEET algorithm achieves a linear speedup for convergence. We further conduct extensive numerical experiments to evaluate the performance of the proposed NET-FLEET algorithm and verify our theoretical findings.

Bonus-Malus Systems (BMSs) are nowadays widely employed in automobile insurance to dynamically adjust a premium based on a customer's claims experience. The intuition behind these posterior ratemaking systems is that as we observe more claiming behavior, we learn more about the underlying risk profile. These systems are therefore a commercially attractive form of experience rating, in which we correct the prior premium for past claims to reflect our updated beliefs about a customer's risk profile. Moreover, they traditionally consider a customer's number of claims irrespective of their sizes and thus implicitly assume independence between the claim counts and sizes (Hey, 1970; Denuit et al., 2007; Boucher and Inoussa, 2014; Verschuren, 2021). Alternative Bayesian forms of experience rating typically depend only on the frequency component as well or consider the two components separately (see, e.g., Denuit and Lang (2004); Bühlmann and Gisler (2005); Mahmoudvand and Hassani (2009); Bermúdez and Karlis (2011, 2017)).

In this work we study high probability bounds for stochastic subgradient methods under heavy tailed noise. In this case the noise is only assumed to have finite variance as opposed to a sub-Gaussian distribution for which it is known that standard subgradient methods enjoys high probability bounds. We analyzed a clipped version of the projected stochastic subgradient method, where subgradient estimates are truncated whenever they have large norms. We show that this clipping strategy leads both to near optimal any-time and finite horizon bounds for many classical averaging schemes. Preliminary experiments are shown to support the validity of the method.

In this paper, we address the problem of conducting statistical inference in settings involving large-scale data that may be high-dimensional and contaminated by outliers. The high volume and dimensionality of the data require distributed processing and storage solutions. We propose a two-stage distributed and robust statistical inference procedures coping with high-dimensional models by promoting sparsity. In the first stage, known as model selection, relevant predictors are locally selected by applying robust Lasso estimators to the distinct subsets of data. The variable selections from each computation node are then fused by a voting scheme to find the sparse basis for the complete data set. It identifies the relevant variables in a robust manner. In the second stage, the developed statistically robust and computationally efficient bootstrap methods are employed. The actual inference constructs confidence intervals, finds parameter estimates and quantifies standard deviation. Similar to stage 1, the results of local inference are communicated to the fusion center and combined there. By using analytical methods, we establish the favorable statistical properties of the robust and computationally efficient bootstrap methods including consistency for a fixed number of predictors, and robustness. The proposed two-stage robust and distributed inference procedures demonstrate reliable performance and robustness in variable selection, finding confidence intervals and bootstrap approximations of standard deviations even when data is high-dimensional and contaminated by outliers.

Human Activity Recognition (HAR) is a problem of interpreting sensor data to human movement using an efficient machine learning (ML) approach. The HAR systems rely on data from untrusted users, making them susceptible to data poisoning attacks. In a poisoning attack, attackers manipulate the sensor readings to contaminate the training set, misleading the HAR to produce erroneous outcomes. This paper presents the design of a label flipping data poisoning attack for a HAR system, where the label of a sensor reading is maliciously changed in the data collection phase. Due to high noise and uncertainty in the sensing environment, such an attack poses a severe threat to the recognition system. Besides, vulnerability to label flipping attacks is dangerous when activity recognition models are deployed in safety-critical applications. This paper shades light on how to carry out the attack in practice through smartphone-based sensor data collection applications. This is an earlier research work, to our knowledge, that explores attacking the HAR models via label flipping poisoning. We implement the proposed attack and test it on activity recognition models based on the following machine learning algorithms: multi-layer perceptron, decision tree, random forest, and XGBoost. Finally, we evaluate the effectiveness of K-nearest neighbors (KNN)-based defense mechanism against the proposed attack.

Recent years have seen significant advances in explainable AI as the need to understand deep learning models has gained importance with the increased emphasis on trust and ethics in AI. Comprehensible models for sequential decision tasks are a particular challenge as they require understanding not only individual predictions but a series of predictions that interact with environmental dynamics. We present a framework for learning comprehensible models of sequential decision tasks in which agent strategies are characterized using temporal logic formulas. Given a set of agent traces, we first cluster the traces using a novel embedding method that captures frequent action patterns. We then search for logical formulas that explain the agent strategies in the different clusters. We evaluate our framework on combat scenarios in StarCraft II (SC2), using traces from a handcrafted expert policy and a trained reinforcement learning agent. We implemented a feature extractor for SC2 environments that extracts traces as sequences of high-level features describing both the state of the environment and the agent's local behavior from agent replays. We further designed a visualization tool depicting the movement of units in the environment that helps understand how different task conditions lead to distinct agent behavior patterns in each trace cluster. Experimental results show that our framework is capable of separating agent traces into distinct groups of behaviors for which our approach to strategy inference produces consistent, meaningful, and easily understood strategy descriptions.

Malware developers use combinations of techniques such as compression, encryption, and obfuscation to bypass anti-virus software. Malware with anti-analysis technologies can bypass AI-based anti-virus software and malware analysis tools. Therefore, classifying pack files is one of the big challenges. Problems arise if the malware classifiers learn packers' features, not those of malware. Training the models with unintended erroneous data turn into poisoning attacks, adversarial attacks, and evasion attacks. Therefore, researchers should consider packing to build appropriate malware classifier models. In this paper, we propose a multi-step framework for classifying and identifying packed samples which consists of pseudo-optimal feature selection, machine learning-based classifiers, and packer identification steps. In the first step, we use the CART algorithm and the permutation importance to preselect important 20 features. In the second step, each model learns 20 preselected features for classifying the packed files with the highest performance. As a result, the XGBoost, which learned the features preselected by XGBoost with the permutation importance, showed the highest performance of any other experiment scenarios with an accuracy of 99.67%, an F1-Score of 99.46%, and an area under the curve (AUC) of 99.98%. In the third step, we propose a new approach that can identify packers only for samples classified as Well-Known Packed.