program synthesis
Automatically generating a new correct program that matches a problem from a few examples; Chollet equates the efficient capacity to do this with general intelligence.
As defined in 280 | François Chollet on Deep Learning and the Meaning of Intelligence
Where it comes up · 3
- Francois Chollet — Why the biggest AI models can't solve simple puzzles9 claims
- Francois Chollet | Why abstraction is the key to intelligence, and what we’re still missing8 claims
- 280 | François Chollet on Deep Learning and the Meaning of Intelligence1 claim
Most important claims · 10 of 18Search all claims for “program synthesis” →
- ARC functions as an IQ test for machine intelligence that, unlike most LLM benchmarks, is designed to resist memorization because each puzzle is novel and requires only 'core knowledge' (elementary physics, objectness, counting) that any young child possesses, rather than vast crammed knowledge.Francois Chollet — Why the biggest AI models can't solve simple puzzles
- General intelligence is the ability to synthesize on the fly a new correct program matching a novel problem from just a few examples, rather than fetching or interpolating memorized programs; if a system could do this efficiently it would have AGI and would also be able to solve ARC, whereas LLMs fail precisely because all they can do is fetch.280 | François Chollet on Deep Learning and the Meaning of Intelligence
- Deep learning (gradient descent) is compute-efficient but data-inefficient and limited to generalizing within its data distribution, while discrete program search (combinatorial search) is extremely data-efficient—learning a generalizable program from one or two examples—but compute-inefficient due to combinatorial explosion; every limitation of one is a strength of the other, so the path forward is to merge them.Francois Chollet — Why the biggest AI models can't solve simple puzzles
- Deep learning is intrinsically limited to local generalization because the substrate of the model is a big parametric curve; to achieve broad or extreme generalization you must move to a different model type—discrete program search/program synthesis—where the model is a discrete graph of operators learned via combinatorial search.Francois Chollet — Why the biggest AI models can't solve simple puzzles
- The ARC dataset deliberately avoids interpolative structure and provides only few examples, making it unsuitable for deep learning; many people have tried deep learning and GPT-3 on ARC with no success, but program synthesis is very effective on ARC.Francois Chollet | Why abstraction is the key to intelligence, and what we’re still missing
- Reasoning has two definitions: (1) identifying a memorized program template, plugging in new values, and running it—which LLMs do—and (2) when no program exists in memory, synthesizing a new program on the fly from bits and pieces of existing programs, which is dramatically harder and what true reasoning requires.Francois Chollet — Why the biggest AI models can't solve simple puzzles
- LLMs are big interpolative memories—parametric curves fitted to a data distribution—that maintain a bank of memorized solution programs; when given a puzzle they fetch and apply the appropriate stored program, which looks like reasoning but is program fetching, not on-the-fly program synthesis.Francois Chollet — Why the biggest AI models can't solve simple puzzles
- LLMs fail ARC not because they cannot parse the small 2D grids (fine-tuning shows they encode solution programs fine for tasks they've seen), but because each new task requires synthesizing a new solution program on the fly, which is precisely what LLMs cannot do—they can only fetch memorized programs.Francois Chollet — Why the biggest AI models can't solve simple puzzles
- Program synthesis is the method to solve reasoning and discrete problems through combinatorial search over graphs of operators from a domain-specific language (DSL); it can solve problems from very few examples (unlike deep learning which requires dense data) but faces the challenge of combinatorial explosion as the set of possible programs grows exponentially with program size.Francois Chollet | Why abstraction is the key to intelligence, and what we’re still missing
- Codex uses language models to interpolate code snippets from Stack Overflow and GitHub; it is not program synthesis but rather an autocomplete tool that can incrementally improve programmer productivity in the same way Gmail autocomplete saves time, but it cannot solve novel problems.Francois Chollet | Why abstraction is the key to intelligence, and what we’re still missing