Yu Feng

The inverse variance–flatness relation in stochastic gradient descent is critical for finding flat minima

2021 · Proceedings of the National Academy of Sciences

Despite tremendous success of the stochastic gradient descent (SGD) algorithm in deep learning, little is known about how SGD finds generalizable solutions at flat minima of the loss function in high-dimensional weight space. Here, we …

Tree Traversal Synthesis Using Domain-Specific Symbolic Compilation

2021 · Zenodo (CERN European Organization for Nuclear Research)

Artifact for ASPLOS 2022 Submission

A Dynamic Recurrent Model for Next Basket Recommendation

2016

Next basket recommendation becomes an increasing concern. Most conventional models explore either sequential transaction features or general interests of users. Further, some works treat users' general interests and sequential behaviors as two totally divided matters, …