Physics & Materials
9 gains0 Traces1 problem
Every measured gain and problem on the record for this topic, newest first.Too few to weigh yet. Updated .
One square per record, darker for stronger evidence:Evidence: 0 high10 moderate0 emerging
10 records so far, too few to weigh yet. Is your study missing? Add it
AI in physics and materials
A Trace is one record with both a gain and a problem, so it counts on both sides: 9 records show a gain and 1 shows a problem. 10 context records are in the record list but not counted, because they report no measured gain or problem.
| Evidence | Gain side | Problem side |
|---|---|---|
| HighStrong study design, backed by at least two independent sources | 0 | 0 |
| ModerateA measured study or an official source | 9 | 1 |
| EmergingEarly, small, unreviewed or reported second-hand | 0 | 0 |
By type of evidence: 8 benchmark or lab tests; 1 controlled comparison; 1 commentary or narrative review. How evidence is scored
10 records so far, too few to weigh yet. A written summary appears once a topic has at least 20 counted records, with at least 5 on each side. Is your study missing? Add it
In a controlled study on one model ductile iron representing heavily contaminated scrap, AI-guided processing raised strength by up to 15.2% and elongation at fracture by up to 73.2%, with an 8.1% cut in manufacturing cost.
In tests on 85,991 simulated transition-metal complexes, an AI model predicted held-out X-ray absorption spectra directly from 3D atomic coordinates, with average R-factors of 0.043 in k-space and 0.019 in R-space.
In benchmark tests, an interpretable AI model matched the best existing precision for predicting nuclear masses, and its learned structure recovered the liquid-drop model and the localized Jaffe factorization principle without being told to.
A review says data scarcity, limited interpretability and a lack of standardized validation still limit reliable real-world use of machine learning for predicting composite properties.
1 record
Published October 17 2025, this peer-reviewed review summarizes how machine learning techniques are being used to predict mechanical behavior of composite materials from experimental and simulation data.
Discover Applied Sciences
ace