Originally published on bloomberg.com
NORTHAMPTON, MA / ACCESSWIRE / May 12, 2023 / The Bloomberg Terminal provides access to more than 35 million financial instruments across all asset classes. That's a lot of data, and to make it useful, AI and machine learning (ML) are playing an increasingly central role in the Terminal's ongoing evolution.
Machine learning is about scouring data at speed and scale that is far beyond what human analysts can do. Then, the patterns or anomalies that are discovered can be used to derive powerful insights and guide the automation of all kinds of arduous or tedious tasks that humans used to have to perform manually.
While AI continues to fall short of human intelligence in many applications, there are areas where it vastly outshines the performance of human agents. Machines can identify trends and patterns hidden across millions of documents, and this ability improves over time. Machines also behave consistently, in an unbiased fashion, without committing the kinds of mistakes that humans inevitably make.
"Humans are good at doing things deliberately, but when we make a decision, we start from whole cloth," says Gideon Mann, Head of ML Product & Research in Bloomberg's CTO Office. "Machines execute the same way every time, so even if they make a mistake, they do so with the same error characteristic."
The Bloomberg Terminal currently employs AI and ML techniques in several exciting ways, and we can expect this practice to expand rapidly in the coming years. The story begins some 20 years ago
Keeping Humans in the Loop
When we started in the 80s, data extraction was a manual process. Today, our engineers and data analysts build, train, and use AI to process unstructured data at massive speeds and scale - so our customers are in the know faster.
The rise of the machines
Prior to the 2000s, all tasks related to data collection, analysis, and distribution at Bloomberg were performed manually, because the technology did not yet exist to automate them. The new millennium brought some low-level automation to the company's workflows, with the emergence of primitive models operating by a series of if-then rules coded by humans. As the decade came to a close, true ML took flight within the company. Under this new approach, humans annotate data in order to train a machine to make various associations based on their labels. The machine "learns" how to make decisions, guided by this training data, and produces ever more accurate results over time. This approach can scale dramatically beyond traditional rules-based programming.
In the last decade, there has been an explosive growth in the use of ML applications within Bloomberg. According to James Hook, Head of the company's Data department, there are a number of broad applications for AI/ML and data science within Bloomberg.
One is information extraction, where computer vision and/or natural language processing (NLP) algorithms are used to read unstructured documents - data that's arranged in a format that's typically difficult for machines to read - in order to extract semantic meaning from them. With these techniques, the Terminal can present insights to users that are drawn from video, audio, blog posts, tweets, and more.
Anju Kambadur, Head of Bloomberg's AI Engineering group, explains how this works:
"It typically starts by asking questions of every document. Let's say we have a press release. What are the entities mentioned in the document? Who are the executives involved? Who are the other companies they're doing business with? Are there any supply chain relationships exposed in the document? Then, once you've determined the entities, you need to measure the salience of the relationships between them and associate the content with specific topics. A document might be about electric vehicles, it might be about oil, it might be relevant to the U.S., it might be relevant to the APAC region - all of these are called topic codes' and they're assigned using machine learning."
All of this information, and much more, can be extracted from unstructured documents using natural language processing models.
Another area is quality control, where techniques like anomaly detection are used to spot problems with dataset accuracy, among other areas. Using anomaly detection methods, the Terminal can spot the potential for a hidden investment opportunity, or flag suspicious market activity. For example, if a financial analyst was to change their rating of a particular stock following the company's quarterly earnings announcement, anomaly detection would be able to provide context around whether this is considered a typical behavior, or whether this action is worthy of being presented to Bloomberg clients as a data point worth considering in an investment decision.
And then there's insight generation, where AI/ML is used to analyze large datasets and unlock investment signals that might not otherwise be observed. One example of this is using highly correlated data like credit card transactions to gain visibility into recent company performance and consumer trends. Another is analyzing and summarizing the millions of news stories that are ingested into the Bloomberg Terminal each day to understand the key questions and themes that are driving specific markets or economic sectors or trading volume in a specific company's securities.
Humans in the loop
When we think of machine intelligence, we imagine an unfeeling autonomous machine, cold and impartial. In reality, however, the practice of ML is very much a team effort between humans and machines. Humans, for now at least, still define ontologies and methodologies, and perform annotations and quality assurance tasks. Bloomberg has moved quickly to increase staff capacity to perform these tasks at scale. In this scenario, the machines aren't replacing human workers; they are simply shifting their workflows away from more tedious, repetitive tasks toward higher level strategic oversight.
"It's really a transfer of human skill from manually extracting data points to thinking about defining and creating workflows," says Mann.
Ketevan Tsereteli, a Senior Researcher in Bloomberg Engineering's Artificial Intelligence (AI) group, explains how this transfer works in practice.
"Previously, in the manual workflow, you might have a team of data analysts that would be trained to find mergers and acquisition news in press releases and to extract the relevant information. They would have a lot of domain expertise on how this information is reported across different regions. Today, these same people are instrumental in collecting and labeling this information, and providing feedback on an ML model's performance, pointing out where it made correct and incorrect assumptions. In this way, that domain expertise is gradually transferred from human to machine."
Humans are required at every step to ensure the models are performing optimally and improving over time. It's a collaborative effort involving ML engineers who build the learning systems and underlying infrastructure, AI researchers and data scientists who design and implement workflows, and annotators - journalists and other subject matter experts - who collect and label training data and perform quality assurance.
"We have thousands of analysts in our Data department who have deep subject matter expertise in areas that matter most to our clients, like finance, law, and government," explains ML/AI Data Strategist Tina Tseng. "They not only understand the data in these areas, but also how the data is used by our customers. They work very closely with our engineers and data scientists to develop our automation solutions."
Annotation is critical, not just for training models, but also for evaluating their performance.
"We'll annotate data as a truth set - what they call a "golden" copy of the data," says Tseng. "The model's outputs can be automatically compared to that evaluation set so that we can calculate statistics to quantify how well the model is performing. Evaluation sets are used in both supervised and unsupervised learning."
Check out "Best Practices for Managing Data Annotation Projects," a practical guide published by Bloomberg's CTO Office and Data department about planning and implementing data annotation initiatives.
READ NOW
View additional multimedia and more ESG storytelling from Bloomberg on 3blmedia.com.
Contact Info:Spokesperson: BloombergWebsite: https://www.3blmedia.com/profiles/bloombergEmail: [emailprotected]
SOURCE: Bloomberg
Link:
Humans in the Loop: AI & Machine Learning in the Bloomberg ... - AccessWire
- New machine-learning algorithms can help optimize the next ... - News-Medical.Net - September 15th, 2023 [September 15th, 2023]
- Meeranda, the Human-Like AI, Welcomes Recognized Machine ... - Canada NewsWire - September 15th, 2023 [September 15th, 2023]
- An Introduction To Diffusion Models For Machine Learning: What ... - Dataconomy - September 15th, 2023 [September 15th, 2023]
- Machine learning improves credit card fraud detection by over 94 ... - Arab News - September 15th, 2023 [September 15th, 2023]
- Machine-learning model predicts CKD progression with 'readily ... - Healio - September 15th, 2023 [September 15th, 2023]
- Machine Learning Operations Market Is Expected to Witness with Strong Growth rate in the forecast period - Benzinga - September 15th, 2023 [September 15th, 2023]
- How machine learning safeguards organizations from modern cyber ... - BetaNews - September 15th, 2023 [September 15th, 2023]
- Yale researchers investigate the future of AI in healthcare - Yale Daily News - September 11th, 2023 [September 11th, 2023]
- Indigenous knowledges informing 'machine learning' could prevent stolen art and other culturally unsafe AI practices - The Conversation Indonesia - September 11th, 2023 [September 11th, 2023]
- Microchip Launches the MPLAB Machine Learning Development Suite for 8-, 16-, 32-Bit MCUs and MPUs - Hackster.io - September 11th, 2023 [September 11th, 2023]
- Learning and predicting the unknown class using evidential deep learning | Scientific Reports - Nature.com - September 11th, 2023 [September 11th, 2023]
- Why Consider Python for Machine Learning and AI? - Analytics Insight - September 11th, 2023 [September 11th, 2023]
- Rise Of The Machine LearningDeep Fakes Could Threaten Our Democracy - Forbes - September 11th, 2023 [September 11th, 2023]
- What is the future of machine learning? - TechTarget - September 11th, 2023 [September 11th, 2023]
- Machine learning-based diagnosis and risk classification of ... - Nature.com - September 11th, 2023 [September 11th, 2023]
- Machine Learning for .NET Developers Starts with ML.NET and ... - Visual Studio Magazine - September 11th, 2023 [September 11th, 2023]
- New Rice Continuing Studies course to explore generative AI ... - Rice News - September 11th, 2023 [September 11th, 2023]
- Scientist in Molecular Engineering by Machine Learning job with ... - Nature.com - September 11th, 2023 [September 11th, 2023]
- Machine learning helps identify metabolic biomarkers that could ... - News-Medical.Net - September 11th, 2023 [September 11th, 2023]
- Amazons Rajeev Rastogi on AI and Machine Learning revolutionising workplace trends - People Matters - September 11th, 2023 [September 11th, 2023]
- Updates on Multitask learning part1(Machine Learning) | by ... - Medium - September 11th, 2023 [September 11th, 2023]
- Google Research Explores: Can AI Feedback Replace Human Input for Effective Reinforcement Learning in Large Language Models? - MarkTechPost - September 11th, 2023 [September 11th, 2023]
- PhD Candidate in Machine Learning in Neurology job with ... - Times Higher Education - September 11th, 2023 [September 11th, 2023]
- Revolutionizing Drug Development with Machine Learning to ... - Cryptopolitan - September 11th, 2023 [September 11th, 2023]
- Prediction of lung papillary adenocarcinoma-specific survival using ... - Nature.com - September 11th, 2023 [September 11th, 2023]
- Seismologists use deep learning to forecast earthquakes - University of California - September 11th, 2023 [September 11th, 2023]
- Stay ahead of the game: The promise of AI for supply chain ... - Washington State Hospital Association - September 11th, 2023 [September 11th, 2023]
- Predicting Stone-Free Status of Percutaneous Nephrolithotomy ... - Dove Medical Press - September 11th, 2023 [September 11th, 2023]
- What is Image Annotation, and Why is it Important in Machine ... - Ground Report - September 11th, 2023 [September 11th, 2023]
- Detection of diabetic patients in people with normal fasting glucose ... - BMC Medicine - September 11th, 2023 [September 11th, 2023]
- Addressing gaps in data on drinking water quality through data ... - Nature.com - September 11th, 2023 [September 11th, 2023]
- Artificial Intelligence: Transforming Healthcare, Cybersecurity, and Communications - Forbes - September 4th, 2023 [September 4th, 2023]
- Machine learning for chemistry: Basics and applications - Phys.org - September 4th, 2023 [September 4th, 2023]
- Harnessing deep learning for population genetic inference - Nature.com - September 4th, 2023 [September 4th, 2023]
- How Apple is already using machine learning and AI in iOS - AppleInsider - September 4th, 2023 [September 4th, 2023]
- Some Experiences Integrating Machine Learning with Vision and ... - Quality Magazine - September 4th, 2023 [September 4th, 2023]
- Here's Why GPUs Are Deep Learning's Best Friend - Hackaday - September 4th, 2023 [September 4th, 2023]
- UWMadison part of effort to advance fusion energy with machine ... - University of Wisconsin-Madison - September 4th, 2023 [September 4th, 2023]
- Machine learning tool simplifies one of the most widely used reactions in the pharmaceutical industry - Phys.org - September 4th, 2023 [September 4th, 2023]
- Revolutionizing Drug Development Through Artificial Intelligence ... - Pharmacy Times - September 4th, 2023 [September 4th, 2023]
- Money, markets and machine learning: Unpacking the risks of adversarial AI - The Hill - September 4th, 2023 [September 4th, 2023]
- Open source in machine learning: experts weigh in on the future - CryptoTvplus - September 4th, 2023 [September 4th, 2023]
- Machine-Learning Tool Sorts Tics From Non-Tics on Video - Medscape - September 4th, 2023 [September 4th, 2023]
- Machine learning and thought, climate impact on health, Alzheimer's ... - Virginia Tech - September 4th, 2023 [September 4th, 2023]
- Machine Learning Regularization Explained With Examples - TechTarget - September 4th, 2023 [September 4th, 2023]
- Advanced Space-led Team Applying Machine Learning to Detect ... - Space Ref - September 4th, 2023 [September 4th, 2023]
- The challenges of reinforcement learning from human feedback (RLHF) - TechTalks - September 4th, 2023 [September 4th, 2023]
- Optimization of therapeutic antibodies for reduced self-association ... - Nature.com - September 4th, 2023 [September 4th, 2023]
- Bayer Is Rapidly Expanding Its Footprint With Artificial Intelligence - Forbes - September 4th, 2023 [September 4th, 2023]
- Machine Learning, Numerical Simulation Integrated To Estimate ... - Society of Petroleum Engineers - September 4th, 2023 [September 4th, 2023]
- 3 Up-and-Coming Machine Learning Stocks to Put on Your Must ... - InvestorPlace - September 4th, 2023 [September 4th, 2023]
- UW-Madison: Cancer diagnosis and treatment could get a boost ... - University of Wisconsin System - September 4th, 2023 [September 4th, 2023]
- Seismologists use deep learning to forecast earthquakes - University of California, Santa Cruz - September 4th, 2023 [September 4th, 2023]
- How Can Hybrid Machine Learning Techniques Help With Effective ... - Dataconomy - September 4th, 2023 [September 4th, 2023]
- Smarter AI: Choosing the Best Path to Optimal Deep Learning - SciTechDaily - September 4th, 2023 [September 4th, 2023]
- This AI Paper Identifies Popular Dynamics in Behavioral and Physiological Smartphone Authentication and their Performance with Various Deep Learning... - September 4th, 2023 [September 4th, 2023]
- Electronic health records and stratified psychiatry: bridge to ... - Nature.com - September 4th, 2023 [September 4th, 2023]
- This AI Research Paper Presents a Comprehensive Survey of Deep Learning for Visual Localization and Mapping - MarkTechPost - September 4th, 2023 [September 4th, 2023]
- Working with Undirected graphs in Machine Learning part2 - Medium - September 4th, 2023 [September 4th, 2023]
- Generative AI at an inflection point: What's next for real-world ... - VentureBeat - September 4th, 2023 [September 4th, 2023]
- Best use cases of t-SNE 2023 part2(Machine Learning) - Medium - August 26th, 2023 [August 26th, 2023]
- Using AI technologies for effective document processing ... - Data Science Central - August 26th, 2023 [August 26th, 2023]
- The Dawn of Intelligence - Embracing AI's Rise and What It Means ... - TechiExpert.com - August 26th, 2023 [August 26th, 2023]
- The TRIPOD-P reporting guideline for improving the integrity and ... - Nature.com - August 22nd, 2023 [August 22nd, 2023]
- What does the future of learning look like? Faculty and students ... - Lumina Foundation - August 22nd, 2023 [August 22nd, 2023]
- $130+ Billion In-Vitro Diagnostics Markets - Global Forecast to 2030 ... - Business Wire - August 22nd, 2023 [August 22nd, 2023]
- How sure is sure? Incorporating human error into machine learning - University of Cambridge news - August 11th, 2023 [August 11th, 2023]
- The 3 Best Machine Learning Stocks to Buy in August - InvestorPlace - August 11th, 2023 [August 11th, 2023]
- Meet Rebecca Gorman, whose company, Aligned AI, is trying to match up human values with machine learning - Fortune - August 11th, 2023 [August 11th, 2023]
- Machine-Learning-Artificial-Intelligence-And-Big-Data-in-the-New ... - Anesthesiology News - August 11th, 2023 [August 11th, 2023]
- Best of artificial intelligence, machine learning will be deployed at Air India; airline is not just anoth - The Economic Times - August 11th, 2023 [August 11th, 2023]
- Machine learning for improved clinical management of cancers of ... - Nature.com - August 11th, 2023 [August 11th, 2023]
- Machine learning model could enable targeted gene therapies for ... - The Hub at Johns Hopkins - August 11th, 2023 [August 11th, 2023]
- Protiviti Achieves AI and Machine Learning in Microsoft Azure ... - PR Newswire - August 11th, 2023 [August 11th, 2023]
- University of North Florida Launches Artificial Intelligence & Machine ... - Fagen wasanni - August 11th, 2023 [August 11th, 2023]
- Guiding Vaccine Development with Machine Learning - HS Today - HSToday - August 11th, 2023 [August 11th, 2023]
- Machine learning and metagenomics reveal shared antimicrobial ... - Nature.com - August 11th, 2023 [August 11th, 2023]
- The Role of Artificial Intelligence and Machine Learning in ... - Fagen wasanni - August 11th, 2023 [August 11th, 2023]
- Unveiling the Power of Classification and Regression in Machine ... - Medium - August 11th, 2023 [August 11th, 2023]
- How to Use Machine Learning to Scale HR Processes - The HR Director Magazine - August 11th, 2023 [August 11th, 2023]