This article in a continuation in my series of articles aimed at'Explainable Artificial Intelligence (XAI)'. If you haven't checked out the first article, I would definitely recommend you to take a quick glance at'Part I -- The Importance of Human Interpretable Machine Learning' which covers the what and why of human interpretable machine learning and the need and importance of model interpretation along with its scope and criteria. In this article, we will be picking up from where we left off and expand further into the criteria of machine learning model interpretation methods and explore techniques for interpretation based on scope. The aim of this article is to give you a good understanding of existing, traditional model interpretation methods, their limitations and challenges. We will also cover the classic model accuracy vs. model interpretability trade-off and finally take a look at the major strategies for model interpretation. Briefly, we will be covering the following aspects in this article. This should get us set and ready for the detailed hands-on guide to model interpretation coming in Part 3, so stay tuned! Model interpretation at heart, is to find out ways to understand model decision making policies better.

Suppressing bones on chest X-rays such as ribs and clavicle is often expected to improve pathologies classification. These bones can interfere with a broad range of diagnostic tasks on pulmonary disease except for musculoskeletal system. Current conventional method for acquisition of bone suppressed X-rays is dual energy imaging, which captures two radiographs at a very short interval with different energy levels; however, the patient is exposed to radiation twice and the artifacts arise due to heartbeats between two shots. In this paper, we introduce a deep generative model trained to predict bone suppressed images on single energy chest X-rays, analyzing a finite set of previously acquired dual energy chest X-rays. Since the relatively small amount of data is available, such approach relies on the methodology maximizing the data utilization. Here we integrate the following two approaches. First, we use a conditional generative adversarial network that complements the traditional regression method minimizing the pairwise image difference. Second, we use Haar 2D wavelet decomposition to offer a perceptual guideline in frequency details to allow the model to converge quickly and efficiently. As a result, we achieve state-of-the-art performance on bone suppression as compared to the existing approaches with dual energy chest X-rays.

Amino acid sequence portrays most intrinsic form of a protein and expresses primary structure of protein. The order of amino acids in a sequence enables a protein to acquire a particular stable conformation that is responsible for the functions of the protein. This relationship between a sequence and its function motivates the need to analyse the sequences for predicting protein functions. Early generation computational methods using BLAST, FASTA, etc. perform function transfer based on sequence similarity with existing databases and are computationally slow. Although machine learning based approaches are fast, they fail to perform well for long protein sequences (i.e., protein sequences with more than 300 amino acid residues). In this paper, we introduce a novel method for construction of two separate feature sets for protein sequences based on analysis of 1) single fixed-sized segments and 2) multi-sized segments, using bi-directional long short-term memory network. Further, model based on proposed feature set is combined with the state of the art Multi-lable Linear Discriminant Analysis (MLDA) features based model to improve the accuracy. Extensive evaluations using separate datasets for biological processes and molecular functions demonstrate promising results for both single-sized and multi-sized segments based feature sets. While former showed an improvement of +3.37% and +5.48%, the latter produces an improvement of +5.38% and +8.00% respectively for two datasets over the state of the art MLDA based classifier. After combining two models, there is a significant improvement of +7.41% and +9.21% respectively for two datasets compared to MLDA based classifier. Specifically, the proposed approach performed well for the long protein sequences and superior overall performance.

The exponential increase in dependencies between the cyber and physical world leads to an enormous amount of data which must be efficiently processed and stored. Therefore, computing paradigms are evolving towards machine learning (ML)-based systems because of their ability to efficiently and accurately process the enormous amount of data. Although ML-based solutions address the efficient computing requirements of big data, they introduce (new) security vulnerabilities into the systems, which cannot be addressed by traditional monitoring-based security measures. Therefore, this paper first presents a brief overview of various security threats in machine learning, their respective threat models and associated research challenges to develop robust security measures. To illustrate the security vulnerabilities of ML during training, inferencing and hardware implementation, we demonstrate some key security threats on ML using LeNet and VGGNet for MNIST and German Traffic Sign Recognition Benchmarks (GTSRB), respectively. Moreover, based on the security analysis of ML-training, we also propose an attack that has a very less impact on the inference accuracy. Towards the end, we highlight the associated research challenges in developing security measures and provide a brief overview of the techniques used to mitigate such security threats.

A basic problem in machine learning is to find a mapping $f$ from a low dimensional latent space to a high dimensional observation space. Equipped with the representation power of non-linearity, a learner can easily find a mapping which perfectly fits all the observations. However such a mapping is often not considered as good as it is not simple enough and over-fits. How to define simplicity? This paper tries to make such a formal definition of the amount of information imposed by a non-linear mapping. This definition is based on information geometry and is independent of observations, nor specific parametrizations. We prove these basic properties and discuss relationships with parametric and non-parametric embeddings.

Deep metric learning aims to learn a deep embedding that can capture the semantic similarity of data points. Given the availability of massive training samples, deep metric learning is known to suffer from slow convergence due to a large fraction of trivial samples. Therefore, most existing methods generally resort to sample mining strategies for selecting nontrivial samples to accelerate convergence and improve performance. In this work, we identify two critical limitations of the sample mining methods, and provide solutions for both of them. First, previous mining methods assign one binary score to each sample, i.e., dropping or keeping it, so they only selects a subset of relevant samples in a mini-batch. Therefore, we propose a novel sample mining method, called Online Soft Mining (OSM), which assigns one continuous score to each sample to make use of all samples in the mini-batch. OSM learns extended manifolds that preserve useful intraclass variances by focusing on more similar positives. Second, the existing methods are easily influenced by outliers as they are generally included in the mined subset. To address this, we introduce Class-Aware Attention (CAA) that assigns little attention to abnormal data samples. Furthermore, by combining OSM and CAA, we propose a novel weighted contrastive loss to learn discriminative embeddings. Extensive experiments on two fine-grained visual categorisation datasets and two video-based person re-identification benchmarks show that our method significantly outperforms the state-of-the-art.

We propose a new optimization method for training feed-forward neural networks. By rewriting the activation function as an equivalent proximal operator, we approximate a feed-forward neural network by adding the proximal operators to the objective function as penalties, hence we call the lifted proximal operator machine (LPOM). LPOM is block multi-convex in all layer-wise weights and activations. This allows us to use block coordinate descent to update the layer-wise weights and activations in parallel. Most notably, we only use the mapping of the activation function itself, rather than its derivatives, thus avoiding the gradient vanishing or blow-up issues in gradient based training methods. So our method is applicable to various non-decreasing Lipschitz continuous activation functions, which can be saturating and non-differentiable. LPOM does not require more auxiliary variables than the layer-wise activations, thus using roughly the same amount of memory as stochastic gradient descent (SGD) does. We further prove the convergence of updating the layer-wise weights and activations. Experiments on MNIST and CIFAR-10 datasets testify to the advantages of LPOM.

Predictive models such as decision trees and neural networks may produce discrimination in their predictions. This paper proposes a method to post-process the predictions of a predictive model to make the processed predictions non-discriminatory. The method considers multiple protected variables together. Multiple protected variables make the problem more challenging than a simple protected variable. The method uses a well-cited discrimination metric and adapts it to allow the specification of explanatory variables, such as position, profession, education, that describe the contexts of the applications. It models the post-processing of predictions problem as a nonlinear optimization problem to find best adjustments to the predictions so that the discrimination constraints of all protected variables are all met at the same time. The proposed method is independent of classification methods. It can handle the cases that existing methods cannot handle: satisfying multiple protected attributes at the same time, allowing multiple explanatory attributes, and being independent of classification model types. An evaluation using four real world data sets shows that the proposed method is as effectively as existing methods, in addition to its extra power.

I am Imtiaz Adam, and this article is an introduction to AI key terminologies and methodologies on behalf of myself and DLS (www.dls.ltd). AI deals with the area of developing computing systems which are capable of performing tasks that humans are very good at, for example recognising objects, recognising and making sense of speech, and decision making in a constrained environment. Narrow AI: the field of AI where the machine is designed to perform a single task and the machine gets very good at performing that particular task. However, once the machine is trained, it does not generalise to unseen domains. This is the form of AI that we have today, for example Google Translate.

The digital telecommunications receiver is an important context for inference methodology, the key objective being to minimize the expected loss function in recovering the transmitted information. For that criterion, the optimal decision is the Bayesian minimum-risk estimator. However, the computational load of the Bayesian estimator is often prohibitive and, hence, efficient computational schemes are required. The design of novel schemes, striking new balances between accuracy and computational load, is the primary concern of this thesis. Two popular techniques, one exact and one approximate, will be studied. The exact scheme is a recursive one, namely the generalized distributive law (GDL), whose purpose is to distribute all operators across the conditionally independent (CI) factors of the joint model, so as to reduce the total number of operators required. In a novel theorem derived in this thesis, GDL, if applicable, will be shown to guarantee such a reduction in all cases. An associated lemma also quantifies this reduction. For practical use, two novel algorithms, namely the no-longer-needed (NLN) algorithm and the generalized form of the Markovian Forward-Backward (FB) algorithm, recursively factorizes and computes the CI factors of an arbitrary model, respectively. The approximate scheme is an iterative one, namely the Variational Bayes (VB) approximation, whose purpose is to find the independent (i.e. zero-order Markov) model closest to the true joint model in the minimum Kullback-Leibler divergence (KLD) sense. Despite being computationally efficient, this naive mean field approximation confers only modest performance for highly correlated models. A novel approximation, namely Transformed Variational Bayes (TVB), will be designed in the thesis in order to relax the zero-order constraint in the VB approximation, further reducing the KLD of the optimal approximation.