Helping computers learn to tackle big-data problems outside their comfort zones
Imagine combing through thousands of mugshots desperately looking for a match. If time is of the essence, the faster you can do this, the better. A*STAR researchers have developed a framework that could help computers learn how to process and identify these images both faster and more accurately1.
Peng Xi of the A*STAR Institute for Infocomm Research notes that the framework can be used for numerous applications, including image segmentation, motion segmentation, data clustering, hybrid system identification and image representation.
A conventional way that computers process data is called representation learning. This involves identifying a feature that allows the program to quickly extract relevant information from the dataset and categorize it — a bit like a shortcut. Supervised and unsupervised learning are two of the main methods used in representation learning. Unlike supervised learning, which relies on costly labeling of data prior to processing, unsupervised learning involves grouping or ‘clustering’ data in a similar manner to our brains, explains Peng.
Subspace clustering is a form of unsupervised learning that seeks to fit each data point into a low-dimensional subspace to find an intrinsic simplicity that makes complex, real-world data tractable. Existing subspace clustering methods struggle to handle ‘out-of-sample’, or unknown, data points and the large datasets that are common today.
“One of the challenges of the big-data era is to organize out-of-sample data using a machine learning model based on ‘in-sample’, or known, observational data,” explains Peng who, with his colleagues, has proposed three methods as part of a unified framework to tackle this issue. These methods differ in how they implement representation learning; one focuses on sparsity, while the other two focus on low rank and grouping effects. “By solving the large-scale data and out-of-sample clustering problems, our method makes big-data clustering and online learning possible,” notes Peng.
The framework devised by the team splits input data into ‘in-sample’ data or ‘out-of-sample’ data during an initial ‘sampling’ step. Next, the in-sample data is grouped into subspaces during the ‘clustering’ step, after which the out-of-sample data is assigned to the nearest subspace. These points are then designated as cluster members.
The team tested their approach on a range of datasets including different types of information, from facial images to text — both handwritten and digital — poker hands and forest coverage. They found that their methods outperformed existing algorithms and successfully reduced the computational complexity (and hence running time) of the task while still ensuring cluster quality.
Learn more: Thinking outside the sample
The Latest on: Big data
[google_news title=”” keyword=”big data” num_posts=”10″ blurb_length=”0″ show_thumb=”left”]
via Google News
The Latest on: Big data
- Data: The Cornerstone Of Smart Citieson May 10, 2024 at 7:15 am
Being a smart city is, after all, not one size fits all, as each has its own approach to becoming a smart city. The goals, however, are consistent.
- How Retail Owners Can Use Data to Big Effecton May 10, 2024 at 12:49 am
With proptech turning a major challenge into a major opportunity, retail owners can unlock a portfolio of possibilities.
- Use Big Data to Revolutionize Women’s Healthon May 9, 2024 at 3:40 am
For far too long, the field of medical research and healthcare has been heavily slanted towards the male physiology as the default study subject. From clinical trials that predominantly used male ...
- Lagrange raises $13M to build blockchain-based cryptographic ‘big data’-scale computationon May 8, 2024 at 11:26 am
Lagrange Labs, a startup building the blockchain-based cryptography protocol Lagrange, today announced it raised $13.2 million in seed funding to progress its vision of scaling up verifiable ...
- Big data helps determine what drives disease riskon May 8, 2024 at 10:21 am
Working with nearly 3,000 observations across almost 1,500 host-parasite combinations, researchers at Notre Dame University have found that biodiversity loss, chemical pollution, introduced species, ...
- Unlocking big data in commercial real estateon May 8, 2024 at 7:00 am
Strategists at a commercial real estate developer recently had a major “Aha!” moment: The markets they are targeting for new projects are precisely those where the competition is most intense, ...
- In an Era of Artificial Intelligence and Big Data, Human Touch Is Neededon May 8, 2024 at 6:19 am
While AI can assist with some decisions, it’s feasible that, at some level, overreliance on these technological capabilities ...
- The power of Big Dataon May 6, 2024 at 8:18 am
Better understand TV audiences with Nielsen’s Big Data + Panel measurement, combining data from digital devices with panel data from actual people.
- Publicis Won’t Use Nielsen’s Big Data as Currency in This Upfronton May 3, 2024 at 6:14 am
Giant media buyer believes numbers haven’t been stable and there isn’t enough historical data to make projections ...
- Big data comes to dinneron May 2, 2024 at 3:30 pm
Artificial intelligence is already changing how people work, communicate online, create art and manage businesses. Now the technology is being used in every aspect of our food systems.
via Bing News