Math: Dynamic programming

Machine-learning ideas tagged Dynamic programming in the Math taxonomy of the Math2NN corpus.

2 ideas found

Solves: Scalability Math: Dynamic programming Clear all

△ Mechanism confirmed, baseline not beaten 2026

Rate-adaptive KV-cache quantizer

Allocate different entropy budgets to different KV-cache blocks instead of assigning every token and head the same nominal bitwidth. Use the ECASQ Lagrangian so high-variance or attention-sensitive blocks receive more codepoints, while predictable blocks collapse to fewer symbols and become highly compressible. Preserve unbiasedness per scalar or block so reconstructed keys and values have zero mean quantization error conditional on the original tensor.

Useful8/10

Difficulty6/10

Novelty7/10

Memory Memory Accuracy Scalability

Paper: Entropy-Constrained Adaptive Stochastic Quantization arXiv:2608.18147

Unverified 2026

Budgeted Tree Message-Passing Planner

Before or during graph-neural-network inference, use a tree dynamic program to choose a limited set of active computation nodes and upgraded message-passing edges. A node receives an embedding from an active landmark only when the selected path has enough residual communication radius, so the planner directly optimizes weighted coverage under a joint node-and-edge budget. The resulting active subgraph is then used by a sparse GNN or graph transformer.

Useful6/10

Difficulty5/10

Novelty7/10

Architecture Speedup Scalability Graph nn

Paper: Maximum Covering Network Design on Graphs with Low Connectivity: Dynamic Programming and Block-Cut Trees arXiv:2608.20894