Yury Yanovich
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Yury Yanovich has a total of 139 co-authors. Here are the top co-authors by number of shared publications:
- Yash Madhwal
Skolkovo Institute of Science and Technology
26 shared publications Grigorii Melnikov
4 shared publications- Vyacheslav A. Davydov
National Research University Higher School of Economics
4 shared publications - Stanislav Kruglik
Technical University of Denmark
3 shared publications - Vladimir Gorgadze
Moscow Institute of Physics and Technology
3 shared publications - Alisa Kalacheva2 shared publications
Latest Publications
- Open Access
- 27 October 2026
- Published by Elsevier in SSRN Electronic Journal
Equal-information evaluation of credit assignment in modular financial reinforcement learning
Credit assignment in modular financial reinforcement learning is difficult to evaluate because an attributed signal may improve a component loss without adding decision-relevant information or improving economic outcomes. We propose an evaluation framework combining equal-information reconstruction controls, fixed-utility comparisons and prespecified downstream guardrails. The framework is instantiated in an attribution-guided modular trading architecture and tested in three controlled studies covering update allocation, supervision-target representation and portfolio risk control. We find that residual-based allocation is exactly reproduced by an ordinary loss-budget control under the disclosed interface; direct shortfall supervision improves component prediction but fails decoded-outcome preservation; and component-wise critic supervision worsens risk-condition certainty equivalent and drawdown relative to scalar supervision under fixed utility. These results identify three distinct evaluation failures: redundant attribution signals, representation-induced prediction trade-offs, and misalignment between component supervision and downstream risk outcomes. A preserved historical engineering case further shows why module-level learning claims require run-level provenance. The contribution is a reusable methodology for evaluating credit assignment in expert systems operating under partial observability, delayed feedback and economic risk constraints.
- 27 October 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessEqual-information evaluation of credit assignment in modular financial reinforcement learning
Credit assignment in modular financial reinforcement learning is difficult to evaluate because an attributed signal may improve a component loss without adding decision-relevant information or improving economic outcomes. We propose an evaluation framework combining equal-information reconstruction controls, fixed-utility comparisons and prespecified downstream guardrails. The framework is instantiated in an attribution-guided modular trading architecture and tested in three controlled studies covering update allocation, supervision-target representation and portfolio risk control. We find that residual-based allocation is exactly reproduced by an ordinary loss-budget control under the disclosed interface; direct shortfall supervision improves component prediction but fails decoded-outcome preservation; and component-wise critic supervision worsens risk-condition certainty equivalent and drawdown relative to scalar supervision under fixed utility. These results identify three distinct evaluation failures: redundant attribution signals, representation-induced prediction trade-offs, and misalignment between component supervision and downstream risk outcomes. A preserved historical engineering case further shows why module-level learning claims require run-level provenance. The contribution is a reusable methodology for evaluating credit assignment in expert systems operating under partial observability, delayed feedback and economic risk constraints.
- Open Access
- 15 August 2026
- Published by Elsevier in SSRN Electronic Journal
Priority Fees and Hedge Design in Cross-Venue Arbitrage between Centralized and Decentralized Exchanges
Arbitrage between centralized (CEX) and decentralized (DEX) exchanges is often modelled as if both legs were executed simultaneously at quoted prices. In practice, the CEX leg can be filled immediately, while the DEX leg waits for block inclusion. We study this mismatch in a model where the trader chooses a priority fee and a linear CEX hedge after observing an arbitrage signal. The priority fee changes the confirmation-time distribution; the hedge changes the exposure during the wait. A mean--variance benchmark gives closed-form rules for both controls: the hedge is a projection on the CEX instrument, while the fee equates marginal gas cost with the value of shorter exposure. A stochastic-control version with execution intensity leads, after a CRRA reduction, to a one-dimensional Hamilton--Jacobi--Bellman equation in the CEX--DEX log spread. Empirical analysis is used strictly to calibrate magnitudes. In low-gas 2026 data, the optimal fee is usually the lower admissible bid. In congested gas regimes, the formula raises the fee because waiting risk and gas costs both enter the same first-order condition.
- 15 August 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessPriority Fees and Hedge Design in Cross-Venue Arbitrage between Centralized and Decentralized Exchanges
Arbitrage between centralized (CEX) and decentralized (DEX) exchanges is often modelled as if both legs were executed simultaneously at quoted prices. In practice, the CEX leg can be filled immediately, while the DEX leg waits for block inclusion. We study this mismatch in a model where the trader chooses a priority fee and a linear CEX hedge after observing an arbitrage signal. The priority fee changes the confirmation-time distribution; the hedge changes the exposure during the wait. A mean--variance benchmark gives closed-form rules for both controls: the hedge is a projection on the CEX instrument, while the fee equates marginal gas cost with the value of shorter exposure. A stochastic-control version with execution intensity leads, after a CRRA reduction, to a one-dimensional Hamilton--Jacobi--Bellman equation in the CEX--DEX log spread. Empirical analysis is used strictly to calibrate magnitudes. In low-gas 2026 data, the optimal fee is usually the lower admissible bid. In congested gas regimes, the formula raises the fee because waiting risk and gas costs both enter the same first-order condition.
- 14 August 2026
- Published by Elsevier in Expert Systems with Applications
- Vol. 333
The bidding games: Reinforcement learning for MEV extraction on Polygon blockchain
- 14 August 2026
- Published by Elsevier in Expert Systems with Applications
- Vol. 333
The bidding games: Reinforcement learning for MEV extraction on Polygon blockchain
- Open Access
- 2 August 2026
- Published by Elsevier in Blockchain: Research and Applications
Breaking Hyperledger’s BFT Shield: Experimental Analysis and Mitigation of Queue Poisoning Attacks in SmartBFT-Go
- 2 August 2026
- Published by Elsevier in Blockchain: Research and Applications
Open AccessBreaking Hyperledger’s BFT Shield: Experimental Analysis and Mitigation of Queue Poisoning Attacks in SmartBFT-Go
- Open Access
- 24 July 2026
- Published by Elsevier in Blockchain: Research and Applications
Fair and Decentralized Committee Selection: Deterministic Bounds for Scalable Distributed Ledgers
- 24 July 2026
- Published by Elsevier in Blockchain: Research and Applications
Open AccessFair and Decentralized Committee Selection: Deterministic Bounds for Scalable Distributed Ledgers
- Open Access
- 3 July 2026
- Published by Elsevier in Blockchain: Research and Applications
DIT: Dimension Reduction View on Optimal NFT Rarity Meters
- 3 July 2026
- Published by Elsevier in Blockchain: Research and Applications
Open AccessDIT: Dimension Reduction View on Optimal NFT Rarity Meters
- 2 July 2026
- Published by Springer Nature
- p. 419-429
Cross-Chain Decentralized KYC Infrastructure for Secure Identity Management in IDOs
- 2 July 2026
- Published by Springer Nature
- p. 419-429
Cross-Chain Decentralized KYC Infrastructure for Secure Identity Management in IDOs
- 2 July 2026
- Published by Springer Nature
- p. 349-360
Multi-Pool Implementation for Peer-to-Peer Lending on Blockchain
- 2 July 2026
- Published by Springer Nature
- p. 349-360
Multi-Pool Implementation for Peer-to-Peer Lending on Blockchain
- 2 July 2026
- Published by Springer Nature
- p. 384-395
Decentralized Salary Payment Automation Using Ethereum Account Abstraction
- 2 July 2026
- Published by Springer Nature
- p. 384-395
Decentralized Salary Payment Automation Using Ethereum Account Abstraction
- 2 July 2026
- Published by Springer Nature
- p. 446-457
Web Application for Generation and Integration of Smart Contracts into Blockchain-Based Supply Chain
- 2 July 2026
- Published by Springer Nature
- p. 446-457
Web Application for Generation and Integration of Smart Contracts into Blockchain-Based Supply Chain