A systems-thinking essay arguing that most optimization quietly trades away buffers, slack, and resilience to make present metrics look better. It reframes efficiency as borrowing stability from the future, and shows how education, workforce, infrastructure, markets, and hardware all get optimized into fragility.

Repository for the study "Predicting the Urban Stormwater Drainage System State using the Graph-WaveNet"

A deep reinforcement learning system for optimizing bridge maintenance decisions across municipal infrastructure fleets, implementing cross-subsidy budget sharing and cooperative multi-agent learning.

Deep Q-Network (DQN) implementation for optimal maintenance planning of 100-bridge fleet infrastructure using advanced reinforcement learning techniques and vectorized parallel training.

Deep Q-Network implementation for optimal bridge maintenance planning using Markov Decision Process formulation with vectorized parallel training. Based on Phase 3 (Vectorized DQN) from dql-maintenance-faster project.

This repository contains data, figures, and analysis supporting SymC Power Grid Optimization, which demonstrates predictive grid stability through scale-invariant substrate inheritance. Using synchrophasor data, the work shows early-warning precursors, irreversible degradation, and control protocols grounded in critical-damping physics.

Local delay-tolerant network system of phones that can be plugged into the internet — like a fallback internet. (PoC)

Global Vulnerability Atlas visualizes planetary risk across climate, ecosystems, infrastructure, conflict, and social stability. A unified dashboard for comparing regional fragility, forecasting cascades, and guiding regenerative, data-driven strategies for a safer future.

This project applies self-improving (Agentic) clustering with Bayesian Optimization to bridge maintenance data in some Prefecture, Japan, to automatically identify bridge groups with high maintenance priority.

This repository analyses landslide events in Turkey from the NASA Global Landslide Catalog using Python, exploring how triggers relate to fatalities and seasonality, and mapping provincial hotspots to inform risk-aware planning.

建設の技術基準に関する質問の専門性粒度（細かい/粗い）を96%正確に自動判定し、最適なRAGシステム（ColBERT/Naive）を選択する実用的なAgentic RAGシステムのMVPです。2025年11月に公開された河川砂防ダムの技術基準を対象に４つのRAGシステムを構築し、専門性の粒度が異なる200問の質問に対して、精度と速度を比較した。

This system analyzes bridge repair method recommendation reports generated by AI agents and visualizes the decision-making pathway from damage → deterioration factors → repair methods as a Decision Tree. It aims to "make the thought process visible."

This tool applies self-improving (Agentic) clustering to bridge maintenance data in Open data at some Prefecture, Japan, to automatically identify bridge groups with high maintenance priority.

⚖️ Explore how optimizing systems can borrow stability from the future, emphasizing resilience and balance over short-term gains.

C51 Distributional DQN (v0.8) for bridge fleet maintenance optimization. Implements categorical return distributions (Bellemare et al., PMLR 2017) with 300x speedup via vectorized projection. Combines Noisy Networks, Dueling DQN, Double DQN, PER, and n-step learning. Validated on 200-bridge fleet: +3,173 reward in 83 min (25k episodes).

Census tract-level earthquake resilience modeling for the City of Los Angeles power grid that compares substation repair prioritization for network connectivity, service restoration, and equity, using network centrality metrics and k-means census-tract resilience typologies.

Gemma3 RAG benchmark system for Japanese river/dam/erosion control technical standards.