XLOG integrates neural perception with logic for faster reasoning
XLOG: A CUDA-Native Engine for Neurosymbolic Integration
Artificial Intelligence
Summary
Combining neural networks that recognize patterns with symbolic logic for reasoning can be complex and slow. The authors describe xlog, a software tool that runs directly on graphics processors to speed up this process. It supports different types of logic reasoning, including exact and probabilistic inference, while reducing the need to move data between the processor and computer memory. Their tests show xlog can speed up training tasks and handle large graph problems more efficiently than some previous tools. However, some reasoning tasks remain slower compared to other specialized systems.
What this means in practice
- •For machine learning engineers: Train models on integrated symbolic and neural data faster using xlog’s GPU-accelerated logic engine.
- •For data engineers: Perform large-scale graph analytics like triangle counting with improved speed and reduced memory using xlog’s optimized joins.
Authors
Levi Dubrovin, Nikita Pospelov, Kirill Sabitov
Abstract
xlog is a CUDA-native logic programming engine integrating neural perception with deterministic Datalog, probabilistic inference, and epistemic world views through a typed frontend and provider-owned CUDA runtime. Its reasoning modes share device data planes, but their execution boundaries differ: ordinary Datalog and exact inference are host-orchestrated, while certified resident recursive and Monte Carlo sampled cores record zero tracked host-device transfers before a bounded terminal receipt. The probabilistic path supports end-to-end gradients through GPU knowledge compilation from provenance to CNF to Decision-DNNF, exact weighted model counting, and backward gradients. A final smoothed circuit is certified against its source formula before caching or evaluation. Circuit caching yields a 2.74x MNIST-addition training speedup; a worst-case-optimal join subsystem yields a 27.96x geometric-mean gain over xlog's binary-join baseline. MNIST-addition accuracy matches Scallop's (0.9561 versus 0.9468), but no per-epoch speed claim is made because baseline epoch time varies with CPU quota. In five hub-skewed triangle-counting cases, the Souffle-to-fused-xlog execution-time ratio rises from 0.88x at 150k edges, where Souffle is faster, to 5.54x at 1.2M; fused peak device allocations are 85-1,033 MB versus 3,287-44,979 MB for the materializing arm. Exact inference is correctness-equivalent to but slower than ProbLog2. On a public video benchmark, a proximity predicate trained only through symbolic credit replaces hand-set geometry at unchanged held-out accuracy; within Event-Calculus rule search it fails ten-fold cross-validation and does not transfer on a leak-free split. On a maritime corpus, weighted clauses beat crisp selection by 0.065 F1, with the result reproduced by one chronological training pass.