Vinod Kumar Chauhan

Agentic AI is the most recent AI paradigm that extends the capabilities of large language models (LLMs) by providing access to external tools for performing specific tasks, external information resources for grounding responses rather than relying solely on model knowledge, and persistent task state and memory. These capabilities allow AI systems to plan, act, and adapt to intermediate results, whilst maintaining progress and coordinating specialised agents, allowing complex tasks to be completed with varying degrees of autonomy. For example, an agentic systematic-review assistant can search the web for relevant papers, screen and analyse them, and synthesise the evidence into a PRISMA-informed review. Research and development in agentic AI have expanded rapidly, accompanied by a growing range of general-purpose and specialised agent systems and development frameworks. Although several recent reviews have examined frameworks, evaluation, taxonomies and applications, practical hands-on guidance for researchers and practitioners seeking to understand and build agentic systems remains limited. This paper addresses this gap by providing a practical, implementation-oriented tutorial for building agentic AI systems using plain Python as well as three representative frameworks: LangGraph, CrewAI and the OpenAI Agents SDK. We discuss the progression from basic LLMs to agentic AI, explain the core components of agentic systems and examine agentic workflows and control patterns through which these components are combined to solve tasks; the tutorial then addresses evaluation at the component, trajectory and task levels, together with robustness, failure recovery and practical design guidance. Finally, we implement a research assistant as a case study to illustrate how agentic AI systems can be built, evaluated and safeguarded, supported by concise code skeletons in the paper and complete runnable implementations and additional end-to-end case studies in the accompanying repository at https://github.com/jmdvinodjmd/agentic-ai-tutorial.

Over the past decade, artificial intelligence (AI) has achieved remarkable breakthroughs in diverse fields, with much of this progress stemming from correlational AI , which harnesses vast datasets and computing to detect patterns and make predictions. Despite these achievements, correlational AI often fails when confronted with distribution shifts, struggles to make predictions under interventions, yields superficial explanations, and can perpetuate biases. These shortcomings highlight the fundamental gap between pattern recognition and causal understanding, limiting the development of robust and responsible AI systems. To address these shortcomings, the emerging field of Causal AI synergises the predictive power of modern AI with causal reasoning, explicitly modelling cause-and-effect relationships through interventions and counterfactuals within formal frameworks like Structural Causal Models or Potential Outcomes . In this paper, we present the first comprehensive review of Causal AI for AI researchers new to causal reasoning, complementing prior reviews focused on causality researchers. We provide a conceptual overview of Causal AI by uniquely structuring this introductory review around four key questions: (1) What are causality and causal AI: an introduction to core concepts and frameworks; (2) Why is causality needed: a discussion of the fundamental limitations of correlational AI; (3) What are the foundational challenges inherent in causal reasoning itself; and (4) What are the key emerging research directions at the intersection of causality and AI. Ultimately, this paper aims to provide the rationale and motivation for the AI community to explore why synergising causality with AI holds substantial promise for developing responsible AI systems.

While machine learning algorithms hold promise for personalised medicine, their clinical adoption remains limited, partly due to biases that can compromise the reliability of predictions. In this article, we focus on sample selection bias (SSB), a specific type of bias where the study population is less representative of the target population, leading to biased and potentially harmful decisions. Despite being well-known in the literature, SSB remains scarcely studied in machine learning for healthcare. Moreover, the existing machine learning techniques try to correct the bias mostly by balancing distributions between the study and the target populations, which may result in a loss of predictive performance. To address these problems, our study illustrates the potential risks associated with SSB by examining SSB’s impact on the performance of machine learning algorithms. Most importantly, we propose a new research direction for addressing SSB, based on the target population identification rather than the bias correction. Specifically, we propose two independent networks (T-Net) and a multitasking network (MT-Net) for addressing SSB, where one network/task identifies the target subpopulation which is representative of the study population and the second makes predictions for the identified subpopulation. Our empirical results with synthetic and semi-synthetic datasets highlight that SSB can lead to a large drop in the performance of an algorithm for the target population as compared with the study population, as well as a substantial difference in the performance for the target subpopulations that are representative of the selected and the non-selected patients from the study population. Furthermore, our proposed techniques demonstrate robustness across various settings, including different dataset sizes, event rates and selection rates, outperforming the existing bias correction techniques.

Hypernetworks, or hypernets for short, are neural networks that generate weights for another neural network, known as the target network. They have emerged as a powerful deep learning technique that allows for greater flexibility, adaptability, dynamism, faster training, information sharing, and model compression. Hypernets have shown promising results in a variety of deep learning problems, including continual learning, causal inference, transfer learning, weight pruning, uncertainty quantification, zero-shot learning, natural language processing, and reinforcement learning. Despite their success across different problem settings, there is currently no comprehensive review available to inform researchers about the latest developments and to assist in utilizing hypernets. To fill this gap, we review the progress in hypernets. We present an illustrative example of training deep neural networks using hypernets and propose categorizing hypernets based on five design criteria: inputs, outputs, variability of inputs and outputs, and the architecture of hypernets. We also review applications of hypernets across different deep learning problem settings, followed by a discussion of general scenarios where hypernets can be effectively employed. Finally, we discuss the challenges and future directions that remain underexplored in the field of hypernets. We believe that hypernetworks have the potential to revolutionize the field of deep learning. They offer a new way to design and train neural networks, and they have the potential to improve the performance of deep learning models on a variety of tasks. Through this review, we aim to inspire further advancements in deep learning through hypernetworks.

A trolley is a container for loading printed circuit board (PCB) components, and a trolley optimisation problem (TOP) is an assignment of PCB components to trolleys for use in the production of a set of PCBs in an assembly line. In this paper, we introduce the TOP, a novel operation research application. To formulate the TOP, we derive a novel extension of the bin packing problem. We exploit the problem structure to decompose the TOP into two smaller, identical, and independent problems. Further, we develop a mixed integer linear programming model to solve the TOP and prove that the TOP is an NP-complete problem. A case study of an aerospace manufacturing company is used to illustrate the TOP which successfully automated the manual process in the company and resulted in significant cost reductions and flexibility in the building process.

Electronic Health Records (EHRs) play a crucial role in shaping predictive are models, yet they encounter challenges such as significant data gaps and class imbalances. Traditional Graph Neural Network (GNN) approaches have limitations in fully leveraging neighbourhood data or demanding intensive computational requirements for regularisation. To address this challenge, we introduce CliqueFluxNet, a novel framework that innovatively constructs a patient similarity graph to maximise cliques, thereby highlighting strong inter-patient connections. At the heart of CliqueFluxNet lies its stochastic edge fluxing strategy — a dynamic process involving random edge addition and removal during training. This strategy aims to enhance the model’s generalisability and mitigate overfitting. Our empirical analysis, conducted on MIMIC-III and eICU datasets, focuses on the tasks of mortality and readmission prediction. It demonstrates significant progress in representation learning, particularly in scenarios with limited data availability. Qualitative assessments further underscore CliqueFluxNet’s effectiveness in extracting meaningful EHR representations, solidifying its potential for advancing GNN applications in healthcare analytics.

Background: Irregular time series (ITS) are common in healthcare as patient data is recorded in an electronic health record (EHR) system as per clinical guidelines/requirements but not for research and depends on a patient’s health status. Due to irregularity, it is challenging to develop machine learning techniques to uncover vast intelligence hidden in EHR big data, without losing performance on downstream patient outcome prediction tasks. Methods: In this paper, we propose Perceiver, a cross-attention-based transformer variant that is computationally efficient and can handle long sequences of time series in healthcare. We further develop continuous patient state attention models, using Perceiver and transformer to deal with ITS in EHR. The continuous patient state models utilise neural ordinary differential equations to learn patient health dynamics, i.e., patient health trajectory from observed irregular time steps, which enables them to sample patient state at any time. Results: The proposed models’ performance on in-hospital mortality prediction task on PhysioNet-2012 challenge and MIMIC-III datasets is examined. Perceiver model either outperforms or performs at par with baselines, and reduces computations by about nine times when compared to the transformer model, with no significant loss of performance. Experiments to examine irregularity in healthcare reveal that continuous patient state models outperform baselines. Moreover, the predictive uncertainty of the model is used to refer extremely uncertain cases to clinicians, which enhances the model’s performance. Code is publicly available and verified at https://codeocean.com/capsule/4587224. Conclusions: Perceiver presents a computationally efficient potential alternative for processing long sequences of time series in healthcare, and the continuous patient state attention models outperform the traditional and advanced techniques to handle irregularity in the time series. Moreover, the predictive uncertainty of the model helps in the development of transparent and trustworthy systems, which can be utilised as per the availability of clinicians.

Graph Neural Networks (GNN) have emerged as a popular and standard approach for learning from graph-structured data. The literature on GNN highlights the potential of this evolving research area and its widespread adoption in real-life applications. However, most of the approaches are either new in concept or derived from specific techniques. Therefore, the potential of more than one approach in hybrid form has not been studied extensively, which can be well utilized for sequenced data or static data together. We derive a hybrid approach based on two established techniques as generalized aggregation networks and topology adaptive graph convolution networks that solve our purpose to apply on both types of sequenced and static nature of data, effectively. The proposed method applies to both node and graph classification. Our empirical analysis reveals that the results are at par with literature results and better for handwritten strokes as sequenced data, where graph structures have not been explored.