How Do Machines ‘Grok’ Data?
Quanta Magazine || By apparently overtraining them, researchers have seen neural networks discover novel solutions to problems.
AI Researchers Present “Average Gradient Outer Product (AGOP)” to Explain how Neural Networks Learn
AIthority || Considered the “black box” of machine learning, AI neural networks can expand the range of innovations and applications in different areas. AGOP is a unifying mathematical mechanism that establishes the core fundamentals explaining how neural networks learn from general machine learning models.
How Do Neural Networks Learn? A Mathematical Formula Explains How They Detect Relevant Patterns
UC San Diego Today || The insights, published in the journal Science, can also be used to make other types of machine learning architectures more effective.
Mechanism for feature learning in neural networks and backpropagation-free machine learning models
Science || Work by TILOS Foundations team member Mikahil Belkin and TILOS graduate students presents a unifying mathematical mechanism called Average Gradient Outer Product (AGOP) that characterizes feature learning in neural networks.
Large language models can do jaw-dropping things. But nobody knows exactly why.
MIT Technology Review || An interview with TILOS Foundations team member Mikhail (Misha) Belkin about one of the biggest scientific puzzles of our time.
UCSD professor speaks on autonomous vehicles for micro-mobility
Notre Dame Observer || Professor Henrik Christensen speaks to the challenges and benefits of using automated vehicles for trips of short to moderate distance (zero to five miles), which account for a staggering 60% of all travel in the United States.
The TILOS AI Institute: Integrating optimization and AI for chip design, networks, and robotics
AI Magazine || Optimization is a universal quest, reflecting the basic human need todo better. Improved optimizations of energy-efficiency, safety, robustness, and other criteria in engineered systems would bring incalculable societal benefits. But, fundamental challenges of scale and complexity keep many such real-world optimization needs beyond reach. TILOS aims to overcome these challenges in three […]
Redefining Math: National University Introduces Bachelor of Science Degree in Mathematics with Data Analytics & Teaching Focus
National University News || Bridging mathematics and technology: National University’s mathematics department incorporates data science courses developed through TILOS into an updated Bachelor of Science degree program.
How symmetry can come to the aid of machine learning
MIT News || Exploiting the symmetry within datasets, MIT researchers show, can decrease the amount of data needed for training neural networks.
How an emerging economy is fueling an academic renaissance
University Business || “The CHIPS ACT [is] a specific example of how we want to retake high-tech manufacturing and put it on American shores so that American workers benefit,” says Jodi Reeves, chair of National University’s Department of Data Science & Analytics. “That’s something that all educational institutions want to support.”