Proof pending. Core topic summary fields are still materializing.
Current research in AI theory is focused on understanding the limitations and capabilities of AI systems through mathematical frameworks and principles. Key areas of exploration include the distinction between agency and intelligence, the nature of uncertainty in AI, and the theoretical foundations of deep learning architectures. These insights are crucial for builders as they navigate the complexities of designing AI systems that can adapt, learn effectively, and operate reliably in dynamic environments. By addressing the geometric constraints of supervised learning and the potential for self-improving AI, researchers aim to create more robust and efficient models that can better handle real-world applications. This foundational work is essential for advancing AI technology and ensuring its alignment with human values and utility.
Topic-specific paper and score movement from the daily diff ledger.
To operate reliably under changing conditions, complex systems require feedback on how effectively they use resources, not just whether objectives are met. Current AI systems process vast information ...
Scientific theory shift in AI agents requires more than fitting equations to data. An artificial scientific agent must detect whether an existing representational framework remains transportable into ...
We prove that empirical risk minimisation (ERM) imposes a necessary geometric constraint on learned representations: any encoder that minimises supervised loss must retain non-zero Jacobian sensitivit...
Transformer networks have achieved remarkable empirical success across a wide range of applications, yet their theoretical expressive power remains insufficiently understood. In this paper, we study t...
The paper investigates whether and how AI systems can realize states of uncertainty. By adopting a functionalist and behavioral perspective, it examines how symbolic, connectionist and hybrid architec...
Collective intelligence emerges across biological, physical, and artificial systems without central coordination, yet a unifying principle governing such behaviour remains elusive. The Free Energy Pri...
Understanding why gradient-based training in deep networks exhibits strong implicit bias remains challenging, in part because tractable singular-value dynamics are typically available only for balance...
Input Convex Neural Networks (ICNNs) are commonly used in a two-stage manner: one first trains a convex network and then minimizes it over its input in a downstream inference problem. Recent second-or...
The dominant discourse on AI limitations frames the boundary of AI capability as a divide between digital tasks (where AI excels) and physical tasks (where embodiment is required). We argue this frami...
One of the bottlenecks on the way towards recursively self-improving systems is the challenge of interestingness: the ability to prospectively identify which tasks or data hold the potential for futur...
Freshness
Canonical route: /topics
Agent Handoff
Canonical ID ai-theory | Route /topic/ai-theory
REST example
curl https://sciencetostartup.com/api/v1/agent-handoff/topic/ai-theoryMCP example
{
"tool": "search_papers",
"arguments": {
"query": "AI Theory",
"cluster": "AI Theory"
}
}source_context
{
"surface": "topic",
"mode": "topic",
"query": "AI Theory",
"normalized_query": "ai-theory",
"route": "/topic/ai-theory",
"paper_ref": null,
"topic_slug": "ai-theory",
"benchmark_ref": null,
"dataset_ref": null
}Use This Via API or MCP
Topic pages bundle paper counts, viability trends, author concentration, and top questions into one canonical surface your agents can reference before they open Signal Canvas or create a workspace.