By Dave DeFusco
A team of Katz School researchers has solved one of the biggest headaches in modern machine learning—how to make AI models that can adapt to new information without needing to start over from scratch—by developing a deep learning system that updates itself using new information about the world without retraining, and does so in a way that’s easier to understand.
Dr. David Li, director of the Katz School’s M.S. in Data Analytics and Visualization, and Ruixin Chen, a student in the M.S. in Artificial Intelligence, presented their work, “Adaptive Deep Learning with Batch Feature Re-Engineering and Differential Dynamical Systems,” in March at IEEE SoutheastCon—a major technology conference in Charlotte, North Carolina.
“At its heart, our new system shows how to build AI that keeps up with real-world changes, like shifting patterns in infectious diseases, evolving customer behaviors or new financial trends,” said Dr. Li, senior author of the study. “Most deep learning models are trained once on a big set of data and then expected to make predictions. But real-world data doesn’t sit still.”
Imagine training a model to predict the spread of a disease based on old data. What happens when the disease mutates, or people’s behaviors change? The model gets worse and worse over time. To fix it, the whole system usually needs to be retrained, which takes a lot of time, money and computing power—and might still miss the mark because modelers are chasing a moving target.
“Even models that try to combine machine learning with the old-school physics of systems modeling, like differential equations, often struggle,” said Chen, lead author of the study. “They are either too rigid, too complicated, too slow or too specific to one problem.”
Dr. Li and Chen’s work answers: What if we could create a smarter way for AI to learn from new information, right as it comes in, without constantly retraining? Their idea rests on the following two breakthroughs.
Batch Feature Re-Engineering: When training a deep learning model, data is usually fed in little chunks called “batches.” In traditional training, these batches treat every sample as if it’s just like the others. Dr. Li and Chen decided to embed “class information”—knowledge about how categories of data are behaving right now—into each batch. For example, if a sudden wave of new infection cases is happening, the model is told about this trend in real time, without having to retrain the whole system.
The second piece of the puzzle, Differential Dynamical Systems, uses the math behind systems that change over time, known as differential equations, to guide the learning process. Instead of only relying on past data to predict the future, the model simulates possible future changes using equations that mirror how systems naturally evolve—whether that’s how a disease spreads or how stock prices fluctuate. By merging deep learning with dynamic system simulation, the model becomes more flexible and accurate at predicting what’s next—even when patterns change unexpectedly.
This approach solves some of the most stubborn problems in machine learning today:
- No constant retraining needed: The model adapts on the fly as new data comes in.
- Better performance over time: Because it’s always adjusting, the model doesn’t get stale or outdated as quickly.
- More understandable: By tying the learning process to real-world mechanisms, like how diseases spread, it’s easier for scientists and policymakers to trust what the model is telling them.
“Our paper shows this system in action by modeling infectious disease outbreaks,” said Chen. “As the infection patterns changed over time, the adaptive model stayed accurate much longer than traditional models, even as new waves and variants appeared.”
The work by Dr. Li and Ruixin Chen lays the groundwork for a new kind of AI—one that evolves naturally as the world changes, without needing constant human intervention. This could be a game-changer in many fields: predicting economic shifts, managing supply chains, tracking environmental changes or even personalizing healthcare treatments as patient conditions evolve.
“By teaching AI to understand change itself, we’re making it more human-like in its ability to adapt,” said Dr Li. “This is essential for building systems that can operate reliably in the real world, not just in a lab.”
