On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,695 papers · 148 categories

Trend · papers per month

count % of corpus

2 results for “neural-adjoint”

Benchmarking deep inverse models over time, and the neural-adjoint methodcs.LG

Deep learning models for inverse problems are evaluated over time.

problem Solving inverse problems for natural systems from measurements.

method Comparing deep learning approaches on benchmark tasks and proposing neural-adjoint method.

result Neural-adjoint method achieves best performance in many scenarios.

inverse problems deep learning benchmarking neural-adjoint natural systems

2020-09-27abs ↗pdf ↗

SNAPO: Smooth Neural Adjoint Policy Optimization for Optimal Control via Differentiable Simulationcs.LG

SNAPO optimizes policies for complex sequential decisions using differentiable simulation.

problem Optimizing policies for high-dimensional, sequential decisions under uncertainty.

method Embeds neural policy in a differentiable simulator, computes gradients efficiently.

result Produces sensitivities at a cost proportional to one reverse pass, regardless of sensitivity count.

reinforcement learning dynamic programming differentiable simulation policy optimization sequential decision making

2026-05-07abs ↗pdf ↗