HSBC has led the development of a new, more accurate method for detecting and predicting extreme weather events using advanced visual and language processing models.
Views
Read our latest views about Innovation, Disruptive Technology and Digital Partnerships.
Harnessing the power of AI to fight financial crime Opens in new window
Artificial intelligence is helping us find and tackle financial crime faster and more thoroughly than ever before, says Jennifer Calvery.
FinTextQA: A Dataset for Long-form Financial Question Answering Opens in new window
HSBC, together with researchers from HKUST and Harvard, introduces the FinTextQA, a comprehensive dataset for long-form financial question answering, and a Retrieval-Augmented Generation-based system, providing a robust evaluation of diverse question types and contexts in finance.
AI Regulations Opens in new window
Policymakers worldwide are devising new regulations, seeking to promote innovation and mitigate potential adverse impacts of generative AI.
Quantum computers’ ‘Chat GPT moment’ is coming Opens in new window
The world is on the cusp of a quantum revolution and we must be ready for the opportunities and threats that will bring, says Philip Intallura.
AI weather forecasting Opens in new window
The increased frequency and intensity of extreme weather events have stressed the importance of weather forecasting. artificial intelligence could raise forecast accuracy, potentially saving lives and significant climate damages.
Hong Kong Leads in Digital Assets Liquidity Opens in new window
The Hong Kong government’s recent HK$6 billion digital green bonds issuance not only strengthens the city’s position as a global financial hub, but also signposts the future of capital markets.
Humans + machine = an ethical AI future Opens in new window
It’s vital that when humans and machines work together, artificial intelligence is used responsibly and ethically, says EJ Achtner.
Funding the future: Biopharma and AI optimism Opens in new window
Venture capital is an important source of funding for potentially transformative technologies. Biopharma and artificial intelligence are among the sectors attracting the most interest.
Digital Horizons report Opens in new window
Working with renowned trend forecasters and interviewing leading technology experts, our Digital Horizons report unravels the four key trends shaping the business landscape between now and 2030.
Taming Binarized Neural Networks and Mixed-Integer Programs Opens in new window
HSBC has devised a new method for training simplified, explainable neural networks, thereby enhancing their applicability.
The power of Digital currency programmability Opens in new window
Bojan Obradovic believes the programmability of digital currency could be one of its key benefits – as well as one of its key risks.
Generative AI – Hype vs Reality Opens in new window
Watch now as Mark McDonald, Head of Data Science and Analytics, and Piers Butler, Head of Global Research take a deep dive into the rapid transformations being driven by generative artificial intelligence (AI).
Quantum Multiple Kernel Learning in Financial Classification Tasks Opens in new window
HSBC is exploring the use of Quantum Multiple Kernel Learning (QMKL) to improve the quality of predictive models in finance, demonstrating its effectiveness in tasks like fraud detection.
AI develops fast Opens in new window
The pace of development in Generative Artificial Intelligence (AI) continues to be incredibly fast. We explore three key recent developments in our latest report.
Software Code Quality Measurement: Implications from Metric Distributions Opens in new window
HSBC's research categorises software code quality metrics and proposes a distribution-based method for consistent evaluation, providing practical implications for better software adoption and code quality measurement.
NCGAN: A Calibrated Oversampling Technique for Fraud Detection Opens in new window
HSBC introduces the Negative Calibrated Generative Adversarial Network (NCGAN) together with researchers from Fudan University, a new data oversampling technique for fraud detection that improves precision and F1 score in machine learning models.
An Adversarial Learning with Sum of Top-K Loss Framework for Credit Card Fraud Detection Opens in new window
HSBC has developed the Adversarial Learning with Sum of Top-K Loss (AST) framework together with researchers from Fudan University, a novel solution to improve credit card fraud detection amidst imbalanced data and adversarial attacks.
Reinforcement Learning for Gate Synthesis in Noisy Quantum Systems Opens in new window
HSBC employs the Deep Deterministic Policy Gradient (DDPG) algorithm for effective gate synthesis in noisy quantum environments, highlighting the potential of reinforcement learning techniques in quantum computing applications.
Configured Quantum Reservoir Computing for Multi-Task Machine Learning Opens in new window
HSBC has participated in developing the Configured Quantum Reservoir Computing, a method that improves machine learning and outperforms traditional approaches, providing highly precise predictions in multiple fields including gene networks and foreign exchange markets.
Unsupervised quantum machine learning for fraud detection Opens in new window
HSBC has developed quantum protocols for anomaly detection, applying them to credit card fraud detection, demonstrating potential advantages over traditional machine learning methods, especially as system size increases.
Integration of Macroeconomic Data into Multi-Asset Allocation with Machine Learning Techniques Opens in new window
HSBC proposes a new approach to predict market returns for multiple assets by using machine learning techniques on macroeconomic data, enhancing the performance of investment strategies.