# Wen-Ting Wang - egpivo.dev > Science is a means whereby learning is achieved. ## About - [Home](https://egpivo.dev/): Science is a means whereby learning is achieved. - [About](https://egpivo.dev/about): Wen-Ting (Joseph) Wang — Machine Learning Engineer & Researcher. Ph.D. in Statistics, production ML, on-chain markets. - [Software](https://egpivo.dev/software): R packages, Python apps, and blockchain tooling. Paper-linked packages are listed under About → Publications. - [Crypto](https://egpivo.dev/crypto): Interactive crypto labs and onchain research prototypes — Uniswap v3, execution lineage, audits, and more. - [Books](https://egpivo.dev/reading-list): Personal reading list of books I recommend or have read — tech and non-tech. ## Articles - [Tokyo Was Closed. A Hyperliquid Perpetual Had Already Priced the Nikkei Gap.](https://egpivo.dev/articles/what-priced-the-yen-shock-while-tokyo-was-closed): A perpetual priced the 31 July move before Tokyo opened. The Nikkei index opening print did not, and the ETF and the shares did. [Markets & Execution; Hyperliquid, RWA, Tokenization] - [What Perpetual Dashboards Don't Tell You](https://egpivo.dev/articles/reading-a-perpetual-market): Evidence from Hyperliquid, GMX, and OKX [Markets & Execution; DeFi, Hyperliquid] - [Staked ETH Beyond Staking: Liquidity, Collateral, and Credit](https://egpivo.dev/articles/three-layers-staking): How staking separates capital from operation—and creates new routes to liquidity and credit. [Protocols & Infrastructure; Ethereum, DeFi, Consensus] - [On-Chain Perpetuals: The Difference Between Virtual Liquidity and Real Capital](https://egpivo.dev/articles/on-chain-perpetuals-virtual-liquidity-and-real-capital): Position size sets dollar PnL. Market structure sets the execution price. Loss-absorbing capital sets which losses can be paid. [Markets & Execution; DeFi, DEX] - [Ethereum Charges for Gas Used. Solana Legacy/v0 Prices Requested Compute.](https://egpivo.dev/articles/ethereum-and-solana-price-different-quantities): Ethereum settles execution against gas used; Solana legacy/v0 calculates native priority fees from requested compute units. [Protocols & Infrastructure; Ethereum, Solana] - [Pool Turnover Isn’t Execution Cost: Pricing a $50,000 Uniswap v3 Swap](https://egpivo.dev/articles/why-high-volume-tvl-doesnt-guarantee-a-good-swap-price): What a fixed-block quote reveals about pool activity, price impact, fees, and slippage tolerance. [Markets & Execution; DEX, Ethereum, Uniswap] - [What Happened to Treasuries, Bitcoin, and Tokenized Funds After 23 Yen Shocks?](https://egpivo.dev/articles/twenty-three-yen-shocks-chain-that-didnt-repeat): Across 23 large USD/JPY moves, U.S. yields, BTC/ETH, and a tokenized-Treasury allocation proxy moved—but not in a stable, repeatable sequence. [Tokenized Assets; RWA, Tokenization] - [Execution Surplus and Welfare in a Simplified Intent-Auction Game](https://egpivo.dev/articles/cow-execution-surplus-social-welfare): Solver competition and endogenous participation shape private execution gains and aggregate welfare across batch durations in a simplified intent-auction game. [Markets & Execution; Ethereum, DEX, MEV] - [TVL, Marginal Price, and Liquidity in an AMM](https://egpivo.dev/articles/tvl-marginal-price-and-liquidity): TVL is a measure of capital. The marginal price is the current quote. Liquidity tells us how much trading it takes to move that quote. [Markets & Execution; DEX, Ethereum, Uniswap] - [What Glamsterdam Would Change About Settling Korean Stocks on Ethereum](https://egpivo.dev/articles/what-glamsterdam-would-change-about-settling-korean-stocks-on-ethereum): A settlement workload calibrated to KRX's daily trading value fits within Ethereum's theoretical block-gas ceiling at the baseline. What changes under Glamsterdam is the cost of updating balances and creating new onchain state. [Protocols & Infrastructure; Ethereum, Tokenization, Smart Contracts, RWA] - [A Solana Program Change Can Be Compatible with One Consumer but Not Another](https://egpivo.dev/articles/compatibility-is-an-edge-not-a-program): Testing Provider Changes Against Consumer-Declared Requirements [Protocols & Infrastructure; Solana, Rust] - [Who Can Liquidate an Onchain Loan? From Health Factor to Transaction Execution](https://egpivo.dev/articles/who-gets-to-liquidate-your-loan): A collateral boundary still exists onchain. The difference is who may submit the liquidation once it is reached. [Markets & Execution; Ethereum] - [What Changes When Tokenization Reaches Market Scale?](https://egpivo.dev/articles/what-happens-when-a-capital-market-moves-onchain): Korea is moving past deposit-token pilots. What changes when tokenization reaches the scale of a capital market. [Tokenized Assets; Tokenization, RWA, Ethereum] - [When an Onchain Opportunity Becomes an Execution Race](https://egpivo.dev/articles/the-rent-is-fixed): When discovery stops being scarce, competition becomes an execution race—and the rule that allocates execution determines how much rent participants keep. [Markets & Execution; MEV] - [How CoW Protocol Uses Auctions to Execute Onchain Trades](https://egpivo.dev/articles/one-swap-eight-solvers-inside-cow-protocols-hidden-auction): CoW batches trades and auctions execution to competing solvers. Its auction data shows what those competing plans look like. [Markets & Execution; DEX, Ethereum] - [When Onchain Price Adjustment and Execution Run on Different Clocks](https://egpivo.dev/articles/two-clocks-on-one-pool): Thirty externally selected Solana price shocks, one Orca pool: sometimes price and landed execution return to normal together. Sometimes they don't. [Markets & Execution; Solana, MEV] - [Testing Compatibility Across Solana Program Changes](https://egpivo.dev/articles/the-program-id-stayed-the-same): Two compiled Solana implementations accepted the same call. Both transactions succeeded; only one preserved the caller’s fixed postcondition. [Protocols & Infrastructure; Solana, Rust] - [When More Traders Stop Adding Much Information](https://egpivo.dev/articles/when-more-traders): A minimal benchmark that separates onchain participation from independent information: what actually moves the price when traders are added. [Markets & Execution; Statistics] - [The Stock Can't Trade Yet. Something Is Already Pricing It.](https://egpivo.dev/articles/the-stock-cant-trade-yet): Four synthetic markets priced companies before their shares could trade. [Markets & Execution; Hyperliquid] - [How Far Does a Market Shock Travel?](https://egpivo.dev/articles/how-far-does-a-market-shock-travel): A bounded event study separates observed trading activity from unproven capital migration after a SpaceX-linked token shock. [Tokenized Assets; RWA, Solana, DeFi] - [The Yen Broke. What Actually Moved Onchain?](https://egpivo.dev/articles/the-yen-broke-what-actually-moved-onchain): When the yen broke, what moved onchain, and in what order? Tracing risk transfer through Hyperliquid and DeFi markets. [Markets & Execution; Hyperliquid] - [Measuring a Commodity Shock in Tokenized Markets with a Rust CLI](https://egpivo.dev/articles/measuring-an-onchain-gold-oil-shock-in-rust): On 8 July 2026, spot gold fell while WTI moved the other way. A Rust CLI measures how that commodity shock showed up in tokenized markets. [Tokenized Assets; RWA, DeFi, Ethereum] - [Markets Are Full of Roads. That Doesn't Mean Capital Takes Them.](https://egpivo.dev/articles/markets-are-full-of-roads): Five Solana wrappers on one company, one issuer-designated conversion route, and nine weeks of swap-level flow through it. [Tokenized Assets; RWA, Solana, DeFi] - [What Can a Landed Solana Transaction Actually Prove About DeFi Execution?](https://egpivo.dev/articles/what-can-a-landed-solana-transaction-prove): Execution provenance reconstructed for all 6,695 landed transactions in a frozen SPYx x OKX router set, and what that evidence can actually prove. [Markets & Execution; Solana] - [When Multiple Pools Behave Like One: Impact-Constrained Capacity Concentration in Uniswap v3](https://egpivo.dev/articles/the-cheapest-trade-is-not-always-the-safe-equilibrium): An empirical note on how multiple Uniswap v3 pools behave like one under an impact-constrained capacity objective. [Markets & Execution; DEX, Ethereum, Uniswap] - [I Searched the Transaction for the Slippage Threshold. I Found the Quote Instead.](https://egpivo.dev/articles/i-searched-the-transaction-for-the-slippage-threshold): A transaction-level audit shows why a quoted slippage threshold can disappear before a Solana swap lands. [Markets & Execution; Solana, Rust] - [Who Owns Execution on Solana?](https://egpivo.dev/articles/who-owns-execution-on-solana): A transaction-level study traces router, provider, and execution-source claims in Solana tokenized-equity trades. [Markets & Execution; Solana] - [Can a Solana Transaction Prove the Router?](https://egpivo.dev/articles/the-interface-says-dflow-can-the-transaction-prove-it): Part II of the DFlow series: what a landed Solana transaction can and cannot prove about which router executed it. [Markets & Execution; Solana, Rust] - [DFlow Shows Where Solana Trading Is Going. Retail Pricing Still Happens at the App.](https://egpivo.dev/articles/different-apps-same-router): As execution becomes an embedded wholesale service, synchronized JTX quotes show that app-level fees still shape the price offered before signing. [Markets & Execution; Solana] - [Before the Result Could Count, the Benchmark Had to Freeze](https://egpivo.dev/articles/before-the-result-could-count-the-benchmark-had-to-freeze): Why the execution benchmark had to be frozen, with data, configuration, and evaluation fixed, before any result could count as evidence. [Markets & Execution; DEX, Ethereum] - [A Large DEX Order Is a Sequential Control Problem](https://egpivo.dev/articles/a-large-order-pays-a-path-not-a-price): A large DEX order pays for a path, not a single price. Framing execution as sequential control over pool state, fees, and liquidity response. [Markets & Execution; DEX, Ethereum] - [How I Turned AMM Logs Into a Causal Pipeline](https://egpivo.dev/articles/frozen-logs-matched-did-r-stack): How frozen AMM logs became a matched difference-in-differences pipeline in R, behind the finding of no large short-run liquidity exit after UNIfication. [Markets & Execution; Statistics, Rust, R] - [When the Fee Switch Turns On, Does Liquidity Walk Away?](https://egpivo.dev/articles/when-the-fee-switch-turns-on-do-lps-walk-away): After the Uniswap fee switch turned on, did liquidity providers leave? A causal look at the short-run liquidity response. [Markets & Execution; DEX, Statistics] - [After Uniswap v4 Went Live. The Pool Is No Longer the Whole Market.](https://egpivo.dev/articles/when-the-pool-gets-options): After Uniswap v4 went live, the pool stopped being the whole market. Mapping where trading and liquidity actually sit now. [Markets & Execution; DEX, Hyperliquid] - [When the Book Keeps Moving. What Hyperliquid Lets Outsiders Reconstruct.](https://egpivo.dev/articles/when-the-book-keeps-moving): Hyperliquid is more than a price story. What an outsider can reconstruct from an onchain order book that keeps moving. [Markets & Execution; DEX, Hyperliquid] - [Dynamic Fees in AMMs: When the Signal Matters](https://egpivo.dev/articles/dynamic-fees-amm-signal-matters): When does a dynamic AMM fee actually help? Simulations of which pool-state signals let a fee price toxic flow. [Markets & Execution; Rust, DEX] - [Hyperliquid Shows What a Market Looks Like Without an AMM Pool](https://egpivo.dev/articles/when-the-market-forms-without-an-amm-pool): When no AMM pool forms, where does the market go? Hyperliquid shows what price formation looks like without one. [Markets & Execution; DEX, RWA, Hyperliquid] - [Before MEV, I Built a Rust AMM Lab to Measure Pool State](https://egpivo.dev/articles/before-mev-build-the-pool): Before studying MEV, a Rust AMM lab that measures pool state directly, so later execution questions start from observable pool mechanics. [Markets & Execution; DEX, RWA, MEV] - [The Token Appeared Twice. The AMM Market Formed Once](https://egpivo.dev/articles/the-same-token-is-not-the-same-market): Canonical deployment tells you which contract is real, not where the market forms. The same token appeared twice; an AMM market formed once. [Tokenized Assets; DeFi, DEX, Solana, RWA] - [The AI Agent Can Act On-Chain. Who Defines Its Authority?](https://egpivo.dev/articles/the-agent-can-act-who-defines-its-authority): The hard question is not whether an AI agent can write to a chain, but who defines its authority and where that boundary is enforced. [Protocols & Infrastructure; RWA] - [When the Swap Actually Lands. What Can an Outsider Reconstruct?](https://egpivo.dev/articles/when-the-swap-actually-lands): A DEX receipt does not explain everything, but it gives outsiders a canonical place to start. [Tokenized Assets; DeFi, RWA, Ethereum] - [When the Quote Becomes Calldata. The Fork Tests Whether It Holds.](https://egpivo.dev/articles/when-the-quote-becomes-a-transaction): A note on turning a Uniswap quote into API-native calldata, then replaying that transaction against pinned mainnet state. [Tokenized Assets; DeFi, RWA, Ethereum] - [The Price Moves First. DEX Routes Fray Before the Exit.](https://egpivo.dev/articles/the-price-moves-first-route-frays-before-exit): Exit risk shows up in the quote before any trade fails: route fragmentation, output deterioration, and shifting summary fields. [Tokenized Assets; DeFi, RWA, Ethereum] - [From RWA Evidence Collectors to a Rust Audit Toolkit](https://egpivo.dev/articles/how-i-built-rwa-audit-in-rust): How separate RWA evidence collectors became one Rust CLI: three contracts, versioned evidence bundles, and reproducible audits. [Tokenized Assets; RWA, DeFi, Rust] - [Where RWA Exchange Risk Actually Sits](https://egpivo.dev/articles/where-rwa-exchange-risk-actually-sits): When the venue becomes part of the product, which record is authoritative? [Tokenized Assets; RWA, DeFi, Tokenization, Solana] - [Building an Interactive P2P Attack Gym in Rust](https://egpivo.dev/articles/making-p2p-attack-metrics-visible): An interactive P2P attack gym in Rust: slot pressure, fake state, and poisoned discovery, made visible on 127.0.0.1. [Protocols & Infrastructure; Rust] - [Where RWA Flow Leaves Traces](https://egpivo.dev/articles/where-rwa-trades-and-exits-actually-clear): When RWA trading becomes a clean service, which parts of the path stop being visible? [Tokenized Assets; RWA, DeFi, Tokenization] - [Three Classic Blockchain Attacks, Rebuilt in a Rust Local Lab](https://egpivo.dev/articles/building-blockchain-p2p-security-gym-rust): Sybil, Eclipse, and network-partition attacks, traced from real incidents and rebuilt as local simulations in a Rust lab. [Protocols & Infrastructure; Rust] - [When Real-World Assets Move On-Chain, What Becomes Visible?](https://egpivo.dev/articles/if-everything-can-be-tokenized-what-should-we-audit): When real-world assets move onchain, the token can move faster than the evidence. What becomes visible, and what an audit should check. [Tokenized Assets; RWA, DeFi, Tokenization] - [Building a Stablecoin Audit Toolkit Around Evidence, Not Dashboards](https://egpivo.dev/articles/stablecoin-audit-software-architecture-problem): A Rust-based pipeline that turns stablecoin transfer checks into artifacts, manifests, and an inspectable evidence console. [Tokenized Assets; DeFi, Stablecoins, Rust] - [The Stablecoin Map: What Crypto's Cash Rails Depend On](https://egpivo.dev/articles/stablecoin-map-local-pegs-dollar-rails): Stablecoins are becoming crypto's cash layer. The question is not only what they are pegged to, but what they depend on. [Tokenized Assets; DeFi, Stablecoins, Ethereum] - [The SpaceX Trade Exists. Now Watch the Tape.](https://egpivo.dev/articles/spacex-trade-watch-the-tape): SpaceX stock is not public yet, but SPCXUSDT perps already show volume, open interest, funding, and venue concentration. [Tokenized Assets; RWA, Tokenization] - [Local Pegs, Dollar Rails: Auditing XSGD and EURC Liquidity](https://egpivo.dev/articles/local-pegs-dollar-rails-geo-stablecoin-audit): Auditing onchain liquidity for the local-currency stablecoins XSGD and EURC, and how much of it still runs on dollar rails. [Tokenized Assets; DeFi, Stablecoins, Ethereum] - [Ethereum Gas Changepoints Are Warning Signals, Not Regime Labels](https://egpivo.dev/articles/gas-fee-change-points): Changepoint detection on Ethereum base fees under EIP-1559: useful as early-warning signals for fee-risk stress tests, not as regime labels. [Markets & Execution; Ethereum, Statistics] - [USDC Shows Why Stablecoin Risk Analysis Is Not One Signal](https://egpivo.dev/articles/usdc-looks-like-one-token): USDC behaves differently across chains, so stablecoin risk can't be read from a single signal. [Tokenized Assets; Stablecoins, Ethereum] - [Tokenization Is Not Just Putting Assets on a Blockchain](https://egpivo.dev/articles/tokenization-is-not-just-putting-assets-on-a-blockchain): Tokenization puts the claim on-chain, but custody, compliance, and settlement still depend on surrounding infrastructure. [Tokenized Assets; Tokenization] - [Funding Is Not Identity: A Conservative Approach to Onchain Entity Linking](https://egpivo.dev/articles/unmasking-did-entity-linking): Funding links are not identity. A conservative onchain entity-linking method, run on a 500+500 Arbitrum governance cohort. [Protocols & Infrastructure; Ethereum] - [Before SNARKs: Building a Mini ZK-Rollup Verifier in Rust](https://egpivo.dev/articles/rust-zkp-a-hand-rolled-mini-rollup): A mini ZK-rollup verifier in Rust, built from the basics most explanations skip: Pedersen commitments, Sigma proofs, and Merkle state. [Protocols & Infrastructure; Rust, Zero-Knowledge] - [ZK State Prune: The Boundaries Behind Rollup Cost Models](https://egpivo.dev/articles/zk-state-prune-guardrails): Not all state behaves the same. The assumptions and guardrails that bound ZK State Prune's rollup cost models. [Protocols & Infrastructure; Zero-Knowledge, Statistics, Go] - [Why ZK Rollups Need a Smarter Way to Manage State](https://egpivo.dev/articles/zk-state-prune): Why ZK rollups need smarter state management: survival models estimate slot reuse to drive cost-aware hot/cold placement. [Protocols & Infrastructure; Zero-Knowledge, Statistics, Go] - [Ethereum Account State: QA Pipeline for a Minimal Token](https://egpivo.dev/articles/ethereum-account-state-qa): The QA pipeline behind a minimal token: testing the contract, off-chain state reconstruction, and the frontend end to end. [Protocols & Infrastructure; Smart Contracts, Solidity, Ethereum] - [Metering Chain: QA That Follows Business Logic](https://egpivo.dev/articles/metering-chain-qa-business-logic): Testing Metering Chain against the guarantees defined in Phases 1-4, with QA cases that follow business logic rather than code paths. [Protocols & Infrastructure; Rust, DePIN] - [Blockchain in 2026: Trust, State, and Why Rust Fits](https://egpivo.dev/articles/from-fundamentals-to-systems-blockchain-rust): Why digital finance still leans on centralized operators, what blockchains change about trust and state, and why Rust fits the systems underneath. [Protocols & Infrastructure; Rust, Consensus] - [Metering Chain Phase 4: Settlement and Economic Finality](https://egpivo.dev/articles/metering-chain-phase4-settlement-finality): From accounting to settlement: how Metering Chain defines when a bill becomes economically final. [Protocols & Infrastructure; Rust, DePIN] - [Metering Chain Phase 3: Capability-Driven Delegation](https://egpivo.dev/articles/metering-chain-phase3-delegation): In DePIN, owners can't sign every event. Capability-driven delegation lets operators submit usage without opening a trust hole. [Protocols & Infrastructure; Rust, DePIN] - [Metering Chain Phase 2: Deterministic Authorization for Multi-Operator DePIN](https://egpivo.dev/articles/metering-chain-phase2-deterministic-auth): Deterministic authorization for multi-operator DePIN: deciding who may submit usage, so replayed history still yields the same bill. [Protocols & Infrastructure; Rust, DePIN] - [Ethereum Account State: A Minimal Token with Reconstructible State](https://egpivo.dev/articles/ethereum-account-state): Can a token's entire state be reconstructed from events alone, without reading storage? A minimal token built to answer that question. [Protocols & Infrastructure; Smart Contracts, Ethereum, Solidity] - [Metering Chain: Deterministic Billing](https://egpivo.dev/articles/metering-chain-deterministic-billing): Replaying usage into deterministic balances (same history, same bill) after one too many "we have the logs, but the bill feels wrong" incidents. [Protocols & Infrastructure; Rust, DePIN] - [Learning Consensus Algorithms: A Hands-On Comparison with Rust](https://egpivo.dev/articles/learning-consensus-algorithms): Five consensus algorithms implemented conceptually in Rust to compare PBFT, Gossip, Eventual Consistency, Quorum-less, and Flexible Paxos. [Protocols & Infrastructure; Rust, Consensus] - [Pay‑As‑You‑Go Demo: The Private Gallery in Web3](https://egpivo.dev/articles/payg-private-gallery-demo): A Private Gallery demo that makes the pay-as-you-go article registries and rental modules visible end to end. [Protocols & Infrastructure; Solidity, Ethereum] - [From Services to Products: Composing Multi-Provider Checkout On-Chain](https://egpivo.dev/articles/payg-pool-protocol-composition-layer): Composing multiple providers into one onchain checkout: a pool layer that turns individual services into products on a pay-as-you-go protocol. [Protocols & Infrastructure; Solidity, Ethereum] - [From PCA to Barlow Twins: A Statistical View of Redundancy Reduction in Self-Supervised Learning](https://egpivo.dev/articles/from-pca-to-barlow-twins-statistical-view-redundancy-reduction): Barlow Twins looks statistical on paper but doesn't behave like PCA in practice. A statistical view of redundancy reduction explains why. [AI & Machine Learning; Statistics] - [Designing Rental Services on a Pay-As-You-Go Protocol](https://egpivo.dev/articles/payg-rental-services): Designing rental services as modules on a pay-as-you-go protocol: the contract design, not a dApp tutorial. [Protocols & Infrastructure; Solidity, Ethereum] - [Building Article Subscription Contracts in Solidity](https://egpivo.dev/articles/payg-service-contracts-article-contracts): Designing the onchain payment and settlement layer for article subscriptions in Solidity. [Protocols & Infrastructure; Solidity, Ethereum] - [Building a Simple Pay-As-You-Go Service in Solidity](https://egpivo.dev/articles/practicing-solidity-transitioning-to-web3): A practice project while learning Solidity: a simple pay-as-you-go service contract. [Protocols & Infrastructure; Solidity, Ethereum] - [Hands-On: `kb-bridge` for Context-Aware Knowledge Base Search](https://egpivo.dev/articles/kb-bridge-hands-on-tutorial): A hands-on tutorial for kb-bridge, the MCP server behind the context-aware retriever, for knowledge-base search. [AI & Machine Learning; NLP, LLM, Python] - [How I'm Building a Context-Aware Retriever to Boost RAG Quality (Part 3: Evaluation)](https://egpivo.dev/articles/context-aware-retriever-mcp-iii): Part 3: evaluating whether the context-aware retriever actually improves RAG answer quality. [AI & Machine Learning; NLP, LLM] - [How I'm Building a Context-Aware Retriever to Boost RAG Quality (Part 2: Implementation)](https://egpivo.dev/articles/context-aware-retriever-mcp-ii): Part 2: implementing the context-aware retriever, from design to working retrieval for knowledge-base question answering. [AI & Machine Learning; NLP, LLM] - [How I’m Building a Context-Aware Retriever to Boost RAG Quality (Part 1: Introduction)](https://egpivo.dev/articles/context-aware-retriever-mcp): Part 1: why and how I designed a context-aware retriever to improve RAG answers for knowledge-base question answering. [AI & Machine Learning; NLP, LLM] - [Hangman with DQN and Transformers](https://egpivo.dev/articles/dqn-for-hangman-game): Training an agent to play Hangman with DQN and Transformers. [AI & Machine Learning; NLP] - [GMV Forecasting via xDeepFM](https://egpivo.dev/articles/gmv-forecasting-via-xdeepfm): A deep-learning proof of concept for GMV forecasting with xDeepFM, with an emphasis on clean, reproducible code. [AI & Machine Learning] - [Decoding LoRA: A Comprehensive Summary on Low-Rank Adaptation](https://egpivo.dev/articles/decoding-lora-a-comprehensive-summary-on-low-rank-adaptation): A summary of LoRA: how low-rank adaptation fine-tunes large language models by training small update matrices instead of the full weights. [AI & Machine Learning; NLP, LLM] - [Comparing Sequence-to-Sequence Decoders: With and Without Attention](https://egpivo.dev/articles/comparing-sequence-to-sequence-decoders-with-and-without-attention): A hands-on comparison of sequence-to-sequence decoders with and without attention, going beyond a standard code walkthrough. [AI & Machine Learning; NLP] - [Note: Derivation of Normal Bayesian Test](https://egpivo.dev/articles/normal-bayesian-test): A derivation of the Normal Bayesian test for hypothesis testing. [Statistics & Spatial Modeling] - [Capturing Dominant Spatial Patterns with Two-Dimensional Locations Using SpatPCA](https://egpivo.dev/articles/SpatPCA-demo-two-dim-location): Using SpatPCA in R to capture the most dominant spatial pattern from data on two-dimensional locations, step by step. [Statistics & Spatial Modeling; R] - [Apply SpatPCA to Capture the Dominant Spatial Pattern with One-Dimensional Locations](https://egpivo.dev/articles/SpatPCA-demo-one-dim-location): A SpatPCA tutorial on recovering the dominant spatial pattern from one-dimensional locations, and how it holds up as the signal-to-noise ratio changes. [Statistics & Spatial Modeling; R] - [個人資料去識別化](https://egpivo.dev/articles/文獻回顧_個資去識別化): 開放資料時代,從蒐集到公開分享,如何在保護個人隱私的同時維持資料可用性:個人資料去識別化的流程與方法。 [Statistics & Spatial Modeling] - [Challenges in EOF Patterns with a Single Variable](https://egpivo.dev/articles/problems_eof_patterns_with_a_single_variable): Potential pitfalls of Empirical Orthogonal Function (EOF) patterns when analyzing a single spatial variable. [Statistics & Spatial Modeling] - [Exploring Dominant Spatial Patterns with a Single Variable](https://egpivo.dev/articles/spatial_patterns_with_a_single_variable): A formal look at exploring the dominant spatial patterns of a single variable, with climate research as the motivating context. [Statistics & Spatial Modeling] - [How to Work on Sea Surface Temperature (SST) Data](https://egpivo.dev/articles/how_to_work_on_SST_data): Step by step: loading and working with sea surface temperature (SST) data in R. [Statistics & Spatial Modeling; R] - [Three Fundamental Aspects of Statistical Models](https://egpivo.dev/articles/Three fundamental aspects of statisitcal models): Notes on Stone's 1985 Annals paper, Additive Regression and Other Nonparametric Models, and the three aspects it uses to frame a statistical model. [Statistics & Spatial Modeling] ## Optional - [RSS feed](https://egpivo.dev/rss.xml): Latest articles as RSS. - [XML sitemap](https://egpivo.dev/sitemap.xml): Complete list of public pages and articles.