Tejas Pradeep Shirodkar
1 indexed paper
Recent (6 mo)
1With code
0Influential cites
0Benchmarked
0Publications per year
126
Top categories
ML×1Differential Geometry×1Optimization and Control×1Stats ML×1
Research Timeline
2026
Dead-Direction Conditioners: Gauge-Equivariant Preconditioning for Deep Networks
This paper introduces DDC, a Dead-Direction Conditioner that keeps a deep network's optimization on the symmetry quotient by conditioning the optimizer's state in the orbit decomposition of a $G$-invariant metric.
Highlighted terms show continued research focus across papers