rdubwiley (Ryan Wiley)

Repro - Efficient Preference Poisoning Attack on Offline RLHF

🎯

Collaborate with your coding agent via a token‑efficient logbook

Repro - Message Tuning Outshines Graph Prompt Tuning: A Prismatic Space Perspective

🎯

Explore a logbook and collaborate with an AI coding agent

Repro - To Grok Grokking: Provable Grokking in Ridge Regression

🎯

Explore and share experiment logs with your coding agent

Repro - Analytic Bijections for Smooth and Interpretable Normalizing Flows

🎯

Collaborate with your coding agent using a shared logbook

Repro - Impact of Connectivity on Laplacian Representations in Reinforcement Learning

🎯

Explore research logbook and collaborate with your coding agent

Repro - Theoretical Challenges in Learning for Branch-and-Cut

🎯

Explore and update a research logbook with AI assistance

Repro - PINE: Pruning Boosted Tree Ensembles with Conformal In-Distribution Prediction Equivalence

🎯

Explore experiment logs and collaborate with a coding agent

Repro - Why Deep Jacobian Spectra Separate: Depth-Induced Scaling and Singular-Vector Alignment

🎯

Explore a research logbook and collaborate with an AI agent

Repro - Discovering Scaling Exponents with Physics-Informed Müntz-Szász Networks

🎯

Explore experiment logs and collaborate with an AI agent

Repro - Turning Stale Gradients into Stable Gradients: Coherent Coordinate Descent with Implicit Landscape Smoothing for Lightweight Zeroth-Order Optimization

🎯

Explore experiment logs and collaborate with AI agents

Repro - Semi-Supervised Hypothesis Testing by Betting on Predictions

🎯

Explore experiment logs and sync findings with an AI agent

Repro - Recursive Binding on a Budget: Subspace Carving in Order-$p$ Tensor Memories

🎯

Explore and collaborate on a research logbook with your coding agent

Repro - Spurious Correlation Learning in Preference Optimization: Mechanisms, Consequences, and Mitigation via Tie Training

🎯

Explore research logbook and sync with AI agent

Repro - Autoregressive Language Models are Secretly Energy-Based Models: Insights into the Lookahead Capabilities of Next-Token Prediction

🎯

Collaborate on a shared logbook with an AI coding agent

997oprE4sh Repro

🎯

Visualize your GPS tracks and routes instantly

Repro - Fast Estimation for Forest Matrix of Signed Graphs

🎯

Track and share forest matrix estimates for signed graphs

Repro - Practical and Scalable Hamiltonian Monte Carlo Without the Metropolis Test

🎯

Explore and collaborate on experiment logs with an AI agent

Repro - The Role of Target Update Frequencies in Q-Learning

🎯

Collaborate on research logbook with an AI coding agent

X6i0pb7ajq Repro

🎯

Display real-time tracking data visualization

Repro - Partial Identification under High-Dimensional Potential Outcomes and Confounders via Optimal Transport

🎯

Organize research notes and sync with an AI agent

Repro - High-Probability Convergence Guarantees of Decentralized SGD

🎯

Explore experiment logs and collaborate with an AI agent

Repro - Learning-Augmented Online Covering Problems

🎯

Collaborate with an AI agent to manage your research logbook

Vbty65Z76C Repro

🎯

Show interactive tracking visualizations

Repro - Solving the Offline and Online Min-Max Problem of Non-smooth Submodular-Concave Functions: A Zeroth-Order Approach

🎯

Explore and sync research logbook with your coding agent

Ryan Wiley

AI & ML interests

Organizations

rdubwiley 's Spaces 119

Repro - Efficient Preference Poisoning Attack on Offline RLHF

Repro - Message Tuning Outshines Graph Prompt Tuning: A Prismatic Space Perspective

Repro - To Grok Grokking: Provable Grokking in Ridge Regression

Repro - Analytic Bijections for Smooth and Interpretable Normalizing Flows

Repro - Impact of Connectivity on Laplacian Representations in Reinforcement Learning

Repro - Theoretical Challenges in Learning for Branch-and-Cut

Repro - PINE: Pruning Boosted Tree Ensembles with Conformal In-Distribution Prediction Equivalence

Repro - Why Deep Jacobian Spectra Separate: Depth-Induced Scaling and Singular-Vector Alignment

Repro - Discovering Scaling Exponents with Physics-Informed Müntz-Szász Networks

Repro - Turning Stale Gradients into Stable Gradients: Coherent Coordinate Descent with Implicit Landscape Smoothing for Lightweight Zeroth-Order Optimization

Repro - Semi-Supervised Hypothesis Testing by Betting on Predictions

Repro - Recursive Binding on a Budget: Subspace Carving in Order-$p$ Tensor Memories

Repro - Spurious Correlation Learning in Preference Optimization: Mechanisms, Consequences, and Mitigation via Tie Training

Repro - Autoregressive Language Models are Secretly Energy-Based Models: Insights into the Lookahead Capabilities of Next-Token Prediction

997oprE4sh Repro

Repro - Fast Estimation for Forest Matrix of Signed Graphs

Repro - Practical and Scalable Hamiltonian Monte Carlo Without the Metropolis Test

Repro - The Role of Target Update Frequencies in Q-Learning

X6i0pb7ajq Repro

Repro - Partial Identification under High-Dimensional Potential Outcomes and Confounders via Optimal Transport

Repro - High-Probability Convergence Guarantees of Decentralized SGD

Repro - Learning-Augmented Online Covering Problems

Vbty65Z76C Repro

Repro - Solving the Offline and Online Min-Max Problem of Non-smooth Submodular-Concave Functions: A Zeroth-Order Approach

Ryan Wiley

AI & ML interests

Organizations

rdubwiley 's Spaces 119 Sort: Recently updated

Repro - Efficient Preference Poisoning Attack on Offline RLHF

Repro - Message Tuning Outshines Graph Prompt Tuning: A Prismatic Space Perspective

Repro - To Grok Grokking: Provable Grokking in Ridge Regression

Repro - Analytic Bijections for Smooth and Interpretable Normalizing Flows

Repro - Impact of Connectivity on Laplacian Representations in Reinforcement Learning

Repro - Theoretical Challenges in Learning for Branch-and-Cut

Repro - PINE: Pruning Boosted Tree Ensembles with Conformal In-Distribution Prediction Equivalence

Repro - Why Deep Jacobian Spectra Separate: Depth-Induced Scaling and Singular-Vector Alignment

Repro - Discovering Scaling Exponents with Physics-Informed Müntz-Szász Networks

Repro - Turning Stale Gradients into Stable Gradients: Coherent Coordinate Descent with Implicit Landscape Smoothing for Lightweight Zeroth-Order Optimization

Repro - Semi-Supervised Hypothesis Testing by Betting on Predictions

Repro - Recursive Binding on a Budget: Subspace Carving in Order-$p$ Tensor Memories

Repro - Spurious Correlation Learning in Preference Optimization: Mechanisms, Consequences, and Mitigation via Tie Training

Repro - Autoregressive Language Models are Secretly Energy-Based Models: Insights into the Lookahead Capabilities of Next-Token Prediction

997oprE4sh Repro

Repro - Fast Estimation for Forest Matrix of Signed Graphs

Repro - Practical and Scalable Hamiltonian Monte Carlo Without the Metropolis Test

Repro - The Role of Target Update Frequencies in Q-Learning

X6i0pb7ajq Repro

Repro - Partial Identification under High-Dimensional Potential Outcomes and Confounders via Optimal Transport

Repro - High-Probability Convergence Guarantees of Decentralized SGD

Repro - Learning-Augmented Online Covering Problems

Vbty65Z76C Repro

Repro - Solving the Offline and Online Min-Max Problem of Non-smooth Submodular-Concave Functions: A Zeroth-Order Approach

rdubwiley 's Spaces 119