Rice U. expert: Key is creating ML systems that question their own predictions
Rice University statistician Genevera Allen says scientists must keep questioning the accuracy and reproducibility of scientific discoveries made by machine-learning techniques until researchers develop new computational systems that can critique themselves.
Allen, associate professor of statistics, computer science and electrical and computer engineering at Rice and of pediatrics-neurology at Baylor College of Medicine, will address the topic in both a press briefing and a general session today at the 2019 Annual Meeting of the American Association for the Advancement of Science (AAAS).
“The question is, ‘Can we really trust the discoveries that are currently being made using machine-learning techniques applied to large data sets?’” Allen said. “The answer in many situations is probably, ‘Not without checking,’ but work is underway on next-generation machine-learning systems that will assess the uncertainty and reproducibility of their predictions.”
Machine learning (ML) is a branch of statistics and computer science concerned with building computational systems that learn from data rather than following explicit instructions. Allen said much attention in the ML field has focused on developing predictive models that allow ML to make predictions about future data based on its understanding of data it has studied.
“A lot of these techniques are designed to always make a prediction,” she said. “They never come back with ‘I don’t know,’ or ‘I didn’t discover anything,’ because they aren’t made to.”
She said uncorroborated data-driven discoveries from recently published ML studies of cancer data are a good example.
“In precision medicine, it’s important to find groups of patients that have genomically similar profiles so you can develop drug therapies that are targeted to the specific genome for their disease,” Allen said. “People have applied machine learning to genomic data from clinical cohorts to find groups, or clusters, of patients with similar genomic profiles.
“But there are cases where discoveries aren’t reproducible; the clusters discovered in one study are completely different than the clusters found in another,” she said. “Why? Because most machine-learning techniques today always say, ‘I found a group.’ Sometimes, it would be far more useful if they said, ‘I think some of these are really grouped together, but I’m uncertain about these others.’”
Learn more: Can we trust scientific discoveries made using machine learning?
The Latest on: Machine-learning discoveries
[google_news title=”” keyword=”machine-learning discoveries” num_posts=”10″ blurb_length=”0″ show_thumb=”left”]
via Google News
The Latest on: Machine-learning discoveries
- Researchers outline promises, challenges of understanding AI for biological discoveryon August 11, 2024 at 10:48 am
Machine learning is a powerful tool in computational biology, enabling the analysis of a wide range of biomedical data such as genomic sequences and biological imaging. But when researchers use ...
- CMU researchers outline promises, challenges of understanding AI for biological discoveryon August 9, 2024 at 10:45 am
Machine learning is a powerful tool in computational biology, enabling the analysis of a wide range of biomedical data such as genomic sequences and biological imaging. But when researchers use ...
- How SMB Law Firms Can Leverage AI In E-Discovery to Generate New Businesson August 9, 2024 at 7:59 am
Machine Learning, Natural Language Processing (NLP), and more sophisticated ... then using that link to upload folders containing the documents to be processed for e-discovery investigation to a ...
- Varonis Announces AI-Powered Data Discovery and Classificationon August 6, 2024 at 8:00 am
Without accurate and complete data classification, it’s impossible to prioritize risk, remediate exposures, or enforce downstream security controls. With the addition of AI classification, Varonis ...
- Novel machine learning-based cluster analysis method that leverages target material propertyon August 6, 2024 at 7:11 am
In materials science, substances are often classified based on defining factors such as their elemental composition or crystalline structure. This classification is crucial for advances in materials ...
- Companies to watch using AI in drug discoveryon August 6, 2024 at 7:00 am
Medical Technology Schools reports that with artificial intelligence, pharmaceutical companies can accelerate the drug discovery process, enhance the precision of targeting specific diseases, and ...
- The Impact of Machine Learning Algorithms on Academic Research Paperson August 5, 2024 at 3:55 pm
Machine learning algorithms have become increasingly prevalent in the field of academic research, revolutionizing the way researchers approach and analyze data. These advanced computational techniques ...
- Perovskite discovery goes automatic: New platform expedites material development for next-gen techon August 5, 2024 at 10:07 am
A new research development, published in Nature Communications, from Queen Mary University of London paves the way for faster discovery of novel perovskite materials with desirable properties for ...
- The Power Of GenAI In Process Discoveryon August 5, 2024 at 5:30 am
The combination of process mining and GenAI is reshaping how companies understand and improve their business processes by combining the precision of machine learning with the nuances of human ...
- Ancient discovery hints Milky Way is billions of years older than we thought: new dataon August 4, 2024 at 9:05 am
Our universe is thought to be about 13.8 billion years old, meaning the Milky Way would have formed during the first billion years of existence as we know it.
via Bing News
