Despite the dynamic nature of the market, large dimensionalities of asset space and varieties of financial products, portfolio optimization is a highly challenging problem. These traditional methods like Markowitz Mean-Variance Optimization and Risk Parity are highly static, and have not been able to adjust to the speed-of-change in market regimes. In this paper, the authors present a new Quantum-Inspired Portfolio Optimization (QIPO-RL) model that combines these interesting approaches to create a framework for a Reinforcement Learning (RL) based dynamic stock allocation algorithm. The framework blends quantum-inspired search techniques, an adaptive RL agent, and asset weights to maximize risk-adjusted asset returns, while maintaining assets in optimal allocation, and provides a way to rebalance a portfolio continuously in response to the changing market environment. The results from experiments were compared with state of the art baselines, which showed that QIPO-RL’s annual return is 16.8%, its Sharpe ratio is 1.61, and the maximum drawdown is 11.5% which is the best among all the competing methods. These findings support the synergism of using a global search method inspired by quantum computers in conjunction with an RL-based adaptation of decisions.

Keywords : Portfolio Optimization, RL, Quantum-Inspired Computing, Dynamic Asset Allocation, Sharpe Ratio.

Authors : Kishore Kumar Sambangi

Title : Quantum-Inspired Portfolio Optimization Using Reinforcement Learning for Dynamic Stock Allocation

Volume/Issue : 2026;3(3 ( July - September ))

Page No : 50 - 56

When it comes to cloud applications, authentication systems are a captive audience of increasingly applicable attacks that steal credentials and vulnerable authentication methods, as well as vulnerable applications that will soon be disrupted by a new problem: quantum computing. Mathematical algorithms such as RSA and Elliptic Curve Cryptography (ECC) algorithms are vulnerable to Shor's algorithm and behavioral solutions do not sufficiently provide the provable security to a regulated cloud environment. The proposed framework, QBAC, is a Quantum-Resistant Behavioral Authentication Framework which combines post-quantum key encapsulation (CRYSTALS-Kyber) with multi-modal behavioral biometric verification and session monitoring in real-time. All these characteristics are combined within a framework that synergizes keystroke, mouse, timing and regularity and feeds a machine-learning based classification model designed to calculate a real-time Authentication Confidence Score (ACS). A Continuous Trust Score (CTS) is being kept in each cloud session that allows for anomaly-based re-authentication before hijack can become successful. Experimental results show that QA performs authentication accurately with 98.4%, False Acceptance Rate is 1.2%, False Rejection Rate is 0.8% and the Quantum Security Strength (QSS) score is 96, outperforming bases of RSA and ECC at all attack complexities and more than traditional Multi-Factor Authentication accuracy, outlier, reject rates and security strengths.

Keywords : Post-Quantum Cryptography; CRYSTALS-Kyber; Behavioural Biometrics; Cloud Authentication

Authors : Lava Kumar Vandrangi

Title : Hybrid Quantum-Resistant Behavioural Authentication Framework for Cloud Systems Using Post-Quantum Cryptography and Behavioural Biometrics

Volume/Issue : 2026;3(3 ( July - September ))

Page No : 9 - 17

Scientific documentation is growing so rapidly that it poses a daunting challenge for scientists to find new applications, new domain links and new fields of study. Even with keyword-based search tools, conventional literature review approaches are inadequate for scaling up to cover large volumes of research and systematically reveal research gaps or grounded hypotheses. This paper introduces AgentSci, a novel multi-agent framework that combines the functionalities of Large Language Models (LLMs), Knowledge Graph Reasoning (KGR) and co-operative multi-agent models to enable automatic scientific discovery pipeline. AgentSci consists of five specialised agents which collaboratively process a corpus of research papers to derive ranked, actionable and novel research hypotheses: Literature Mining Agent, Knowledge Graph Builder, Gap Discovery Agent, Hypothesis Generation Agent and a Validation Agent. To measure the quality and usefulness of the outputs generated, we introduce a composite scoring model that combines five different scores: Knowledge Relevance Score (KRS), Research Gap Score (RGS), Hypothesis Novelty Index (HNI), Agent Consensus Score (ACS), and Scientific Discovery Potential (SDP). The accuracy in gap detection, novelty of hypothesis, and relevance of the research are significantly improved by AgentSci over the conventional retrieval based systems in four scientific research areas demonstrated through experiments. AgentSci is a step towards fully automatic scientific discovery, brought by AI.

Keywords : Multi-Agent Systems, LLM, Knowledge Graph Reasoning, Scientific Discovery, Hypothesis Generation.

Authors : Sriharsha Chollangi

Title : AgentSci: A Multi-Agent Scientific Discovery Framework Using Knowledge Graph Reasoning and Large Language Models

Volume/Issue : 2026;3(3 ( July - September ))

Page No : 1 - 8