Browse by topic
- LLM (9)
- AI Agents (9)
- Domain LLMs (1)
- Math (6)
- Analysis (1)
- Algebra & Foundations (0)
- Geometry & Topology (1)
- Probability & Finance (5)
- Physics (5)
- Reading (4)
- Books (0)
- Papers (4)
- Robotics & Embodied AI (2)
- Reinforcement Learning (2)
- Conferences (1)
- ICML 2026 (1)
-
Anshu & Arunachalam — and where RL-ES quantum state estimation lives inside it
Reading-archive entry for the 2023 survey on complexity of learning quantum states (arXiv 2305.20069). Used as the anchor for my RL-ES quantum-state-estimation drafts (Wasserstein gradient flow + λ-convex exponential decay) and to surface the PL-condition pivot once ES is dropped. The post is split by a hard divider into Direct (what the drafts and advisor feedback commit to — locked until next iteration) and Indirect (my own extensions from this reading — subject to revision).
-
Enhanced POET — and the architecture of open-ended intelligence
Reading-archive entry on Wang et al. (ICML 2020). The paper makes a precise claim about open-ended search; this entry extracts that claim and uses it to anchor a five-axis framework — Discovery × Representation × Selection × Transfer × Memory × Meta — that names what the rest of the field is each (separately) optimizing.
-
The agent loop: a field guide to agentic AI architectures
A survey-first deep dive into what an AI agent actually is — the loop, its seven sub-problems, the design space for each, and where the open research lives. The field first; my own work only at the end.
-
Monte Carlo for risk estimation: from sampling to quantum
A risk-analysis survey of how Monte Carlo turned risk into something estimable rather than guessed at — from the Metropolis–Ulam estimator through CVaR and rare-event variance reduction, then where amortized variational methods (VAE) and quantum amplitude estimation actually fit. Field-first; thirteen 5-field reference cards at the end.