AI is emerging as an efficient companion in medicine. While AI holds promise for reducing the cognitive load of researchers and practitioners, its adoption is often hindered by a lack of trust in new AI advancements. We present sophisticated techniques for developing trustworthy artificial intelligence (AI) models in medicine, bridging breakthroughs in AI research with practical healthcare applications. We will discuss in-depth the four stages (Design, Development, Implementation, and Evaluation) involved in the process of building trustworthy AI models customized for the medical domain. We present various techniques for incorporating important Trustworthy AI principles like data privacy, robustness, explainability, interpretability, medical experts-in-the-loop, and risk assessment while developing AI models for medicine. In contrast to prior tutorials, we make the following two key contributions: (i) While explaining the 'Implementation' stage, we cover various real-world healthcare applications developed as part of research projects in academia in collaboration with medical schools in India and Germany. (ii) By including a health informatics professional as one of the tutorial organizers, we provide a fresh and much-needed perspective on the research challenges and mitigation strategies in building AI models for medicine.

Large-scale language models such as DNABert and LOGO aim to learn optimal gene representations and are trained on the entire Human Reference Genome. However, standard tokenization schemes involve a simple sliding window of tokens like k-mers that do not leverage any gene-based semantics and thus may lead to (trivial) masking of easily predictable sequences, and subsequently inefficient Masked Language Modeling (MLM) training. Therefore, we propose a novel masking algorithm, GENEMASK, for MLM training of gene sequences, where we randomly identify positions in a gene sequence as mask centers and locally select the span around the mask center with the highest Normalized Pointwise Mutual Information (NPMI) to mask. We observe that in the absence of human-understandable semantics in the genomics domain (in contrast, semantic units like words and phrases are inherently available in NLP), GENEMASK-based models substantially outperform the SOTA models (DNABert and LOGO) over four benchmark gene sequence classification datasets in five few-shot settings (10 to 1000-shot). More significantly, the GENEMASK-based DNABert model is trained for less than one-tenth of the number of epochs of the original SOTA model. We also observe a strong correlation between top-ranked PMI tokens and conserved DNA sequence motifs, which may indicate the incorporation of latent genomic information. The codes (including trained models) and datasets are made publicly available at https://github.com/roysoumya/GeneMask.

Abstract Pervasiveness of tracking devices and enhanced availability of spatially located data has deepened interest in using them for various policy interventions, through computational data analysis tasks such as spatial hot spot detection. In this paper, we consider, for the first time to our best knowledge, fairness in detecting spatial hot spots. We motivate the need for ensuring fairness through statistical parity over the collective population covered across chosen hot spots. We then characterize the task of identifying a diverse set of solutions in the noteworthiness-fairness trade-off spectrum, to empower the user to choose a trade-off justified by the policy domain. Being a novel task formulation, we also develop a suite of evaluation metrics for fair hot spots, motivated by the need to evaluate pertinent aspects of the task. We illustrate the computational infeasibility of identifying fair hot spots using naive and/or direct approaches and devise a method, codenamed FiSH, for efficiently identifying high-quality, fair and diverse sets of spatial hot spots. FiSH traverses the tree-structured search space using heuristics that guide it towards identifying noteworthy and fair sets of spatial hot spots. Through an extensive empirical analysis over a real-world dataset from the domain of human development, we illustrate that FiSH generates high-quality solutions at fast response times. Towards assessing the relevance of FiSH in real-world context, we also provide a detailed discussion of how it could fit within the current practice of hot spots policing, as read within the historical context of the evolution of the practice.

We consider the task of learning distributed representations for arithmetic word problems. We outline the characteristics of the domain of arithmetic word problems that make generic text embedding methods inadequate, necessitating a specialized representation learning method to facilitate the task of retrieval across a wide range of use cases within online learning platforms. Our contribution is two-fold; first, we propose several 'operators' that distil knowledge of the domain of arithmetic word problems and schemas into word problem transformations. Second, we propose a novel neural architecture that combines LSTMs with graph convolutional networks to leverage word problems and their operator-transformed versions to learn distributed representations for word problems. While our target is to ensure that the distributed representations are schema-aligned, we do not make use of schema labels in the learning process, thus yielding an unsupervised representation learning method. Through an evaluation on retrieval over a publicly available corpus of word problems, we illustrate that our framework is able to consistently improve upon contemporary generic text embeddings in terms of schema-alignment.