atlas

Research map / Learning theory and algorithms / Combinatorial and search optimization

Learning Differentiable Programs with Admissible Neural Heuristics

NeurIPS 2020 · Learning theory and algorithms › Combinatorial and search optimization · approach: Search for program synthesis · 6 citations (OpenAlex)

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use NEAR-guided IDDFS instead of top-down enumerationenumeration · CRIM13-sniffF1: .221 -> .446Helped
use NEAR-guided IDDFS instead of genetic algorithmgenetic algorithm · Fly-vs.-FlyF1: .868 -> .892Helped
use NEAR-guided A* instead of top-down enumerationenumeration · Bball-ballhandlerF1: .857 -> .918Helped
use NEAR-guided search instead of genetic algorithmgenetic algorithm · CRIM13-otherF1: .768 -> .768No clear effect
use NEAR-guided search instead of MCTSMCTS · Bball-ballhandlerF1: .729 -> .903Helped
use NEAR-guided A* instead of enumerationenumeration · CRIM13-sniffaccuracy: .851 -> .821Hurt
increase structural cost weight lambda from 1 to 8 (ablation)A*-NEAR · CRIM13-sniffF1: 0.46 -> 0.22Hurt
compare NEAR programs with RNNRNN · Fly-vs.-FlyF1: .964 -> .892Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy