Beyond Statistical Learning: Exact Learning Is Essential for General Intelligence
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866916817455284224 |
|---|---|
| author | György, András Lattimore, Tor Lazić, Nevena Szepesvári, Csaba |
| author_facet | György, András Lattimore, Tor Lazić, Nevena Szepesvári, Csaba |
| contents | Sound deductive reasoning -- the ability to derive new knowledge from existing facts and rules -- is an indisputably desirable aspect of general intelligence. Despite the major advances of AI systems in areas such as math and science, especially since the introduction of transformer architectures, it is well-documented that even the most advanced frontier systems regularly and consistently falter on easily-solvable deductive reasoning tasks. Hence, these systems are unfit to fulfill the dream of achieving artificial general intelligence capable of sound deductive reasoning. We argue that their unsound behavior is a consequence of the statistical learning approach powering their development. To overcome this, we contend that to achieve reliable deductive reasoning in learning-based AI systems, researchers must fundamentally shift from optimizing for statistical performance against distributions on reasoning problems and algorithmic tasks to embracing the more ambitious exact learning paradigm, which demands correctness on all inputs. We argue that exact learning is both essential and possible, and that this ambitious objective should guide algorithm design. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_23908 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Beyond Statistical Learning: Exact Learning Is Essential for General Intelligence György, András Lattimore, Tor Lazić, Nevena Szepesvári, Csaba Artificial Intelligence Machine Learning Sound deductive reasoning -- the ability to derive new knowledge from existing facts and rules -- is an indisputably desirable aspect of general intelligence. Despite the major advances of AI systems in areas such as math and science, especially since the introduction of transformer architectures, it is well-documented that even the most advanced frontier systems regularly and consistently falter on easily-solvable deductive reasoning tasks. Hence, these systems are unfit to fulfill the dream of achieving artificial general intelligence capable of sound deductive reasoning. We argue that their unsound behavior is a consequence of the statistical learning approach powering their development. To overcome this, we contend that to achieve reliable deductive reasoning in learning-based AI systems, researchers must fundamentally shift from optimizing for statistical performance against distributions on reasoning problems and algorithmic tasks to embracing the more ambitious exact learning paradigm, which demands correctness on all inputs. We argue that exact learning is both essential and possible, and that this ambitious objective should guide algorithm design. |
| title | Beyond Statistical Learning: Exact Learning Is Essential for General Intelligence |
| topic | Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2506.23908 |