Unverified50% confidenceFactExact time
E(3)-equivariant graph neural networks have demonstrated significant advantages in protein structure prediction and molecular property prediction.
1
Sources
50%
Confidence
Long-term
Relevance
9/2/2026
First Seen
Sources
Related Entities
Related Claims
Unverifiede3nn已被应用于分子动力学势能面拟合（如NequIP、MACE等模型）、蛋白质结构预测和材料性质预测80% similarUnverifiedAlphaFold2 elevated the prediction accuracy of protein monomer structures to near-experimental levels by incorporating attention mechanisms (the Evoformer module) and equivariant neural networks70% similarUnverified分子天然具有SE(3)/E(3)等变性，即物理性质不随整体旋转和平移改变；几何深度学习将物理对称先验内嵌于网络结构，使模型参数量更小、泛化更强65% similarUnverifiedThe phenomenon that optimization of over-parameterized neural networks easily converges to good solutions can be explained by Neural Tangent Kernel theory63% similarUnverified卷积神经网络的核心创新在于权重共享与局部连接机制，通过滑动卷积核共享同一组参数降低模型参数量61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/846462API
curl https://kongchang.com/api/v1/knowledge/claims/846462MCP
get_claim(id=846462)