# Logical Induction

arXiv: https://arxiv.org/abs/1609.03543 · HTML: https://mlsanity.com/paper/1609.03543/

## Key facts

- Cited by 16 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 12 Sep 2016.

## Built on by (most cited first, 16 of 16)

- [Eliciting Latent Predictions from Transformers with the Tuned Lens](https://mlsanity.com/paper/2303.08112/) (2023)
- [Scalable agent alignment via reward modeling: a research direction](https://mlsanity.com/paper/1811.07871/) (2018)
- [The Alignment Problem from a Deep Learning Perspective](https://mlsanity.com/paper/2209.00626/) (2022)
- [Evaluating Superhuman Models with Consistency Checks](https://mlsanity.com/paper/2306.09983/) (2023)
- [Consistency Checks for Language Model Forecasters](https://mlsanity.com/paper/2412.18544/) (2024)
- [Experts Don't Cheat: Learning What You Don't Know By Predicting Pairs](https://mlsanity.com/paper/2402.08733/) (2024)
- [Self-Improvements in Modern Agentic Systems: A Survey](https://mlsanity.com/paper/2607.13104/) (2026)
- [From AGI to ASI](https://mlsanity.com/paper/2606.12683/) (2026)
- [Revealed Rationality: Label-Free Evaluation and Regularization from Representation Theorems](https://mlsanity.com/paper/2608.05015/) (2026)
- [Smart Proofs via Smart Contracts: Succinct and Informative Mathematical Derivations via Decentralized Markets](https://mlsanity.com/paper/2102.03044/) (2021)
- [Temporal Inference with Finite Factored Sets](https://mlsanity.com/paper/2109.11513/) (2021)
- [Market-based Architectures in RL and Beyond](https://mlsanity.com/paper/2503.05828/) (2025)
- [Arguments about Highly Reliable Agent Designs as a Useful Path to Artificial Intelligence Safety](https://mlsanity.com/paper/2201.02950/) (2022)
- [Towards a Law of Iterated Expectations for Heuristic Estimators](https://mlsanity.com/paper/2410.01290/) (2024)
- [Proof-Carrying Cognition: Closing the Verification Gap with Reality-Settled Reward](https://mlsanity.com/paper/2609.09776/) (2026)
- [Math for AI safety: an invitation for mathematicians](https://mlsanity.com/paper/2609.15289/) (2026)

## Abstract (arXiv)

We present a computable algorithm that assigns probabilities to every logical statement in a given formal language, and refines those probabilities over time. For instance, if the language is Peano arithmetic, it assigns probabilities to all arithmetical statements, including claims about the twin prime conjecture, the outputs of long-running computations, and its own probabilities. We show that our algorithm, an instance of what we call a logical inductor, satisfies a number of intuitive desiderata, including: (1) it learns to predict patterns of truth and falsehood in logical statements, often long before having the resources to evaluate the statements, so long as the patterns can be written down in polynomial time; (2) it learns to use appropriate statistical summaries to predict sequences of statements whose truth values appear pseudorandom; and (3) it learns to have accurate beliefs about its own current beliefs, in a manner that avoids the standard paradoxes of self-reference. For example, if a given computer program only ever produces outputs in a certain range, a logical inductor learns this fact in a timely manner; and if late digits in the decimal expansion of π are difficult to predict, then a logical inductor learns to assign ≈ 10% probability to "the nth digit of π is a 7" for large n. Logical inductors also learn to trust their future beliefs more than their current beliefs, and their beliefs are coherent in the limit (whenever φ implies psi, mathbbP_∞(φ) ≤ mathbbP_∞(psi), and so on); and logical inductors strictly dominate the universal semimeasure in the limit. These properties and many others all follow from a single logical induction criterion, which is motivated by a series of stock trading analogies. Roughly speaking, each logical sentence φ is associated with a stock that is worth 1 per share if [...]