Hong Seong Pyo

SabaPivot

·

AI & ML interests

AI Safety

Recent Activity

updated a Space about 1 hour ago

SabaPivot/repro-near-optimal-and-efficient-first-order-algorithm-for-multi-task-learning-with-shared-linea

updated a Space about 1 hour ago

SabaPivot/repro-randomized-feasibility-methods-for-constrained-optimization-with-adaptive-step-sizes

updated a Space about 1 hour ago

SabaPivot/repro-decision-tree-learning-on-product-spaces

View all activity

Organizations

SabaPivot 's Spaces 141

Reproduction: Near-Optimal and Efficient First-Order Algorithm for Multi-Task Learning with Shared Linear Representation

Explore and track research code, traces, and workspace

Reproduction: Randomized Feasibility Methods for Constrained Optimization with Adaptive Step Sizes

Explore experiment logs, traces, and workspace in a web UI

Reproduction: Decision Tree Learning on Product Spaces

Explore code, traces, and workspace logs

Reproduction: Efficient privacy loss accounting for subsampling and random allocation

Explore code, traces, and workspace in an interactive logbook

Reproduction: Parameter-free Dynamic Regret: Time-varying Movement Costs, Delayed Feedback, and Memory

Explore code, traces, and workspace in an interactive logbook

Reproduction: Improved Dimension Dependence for Bandit Convex Optimization with Gradient Variation

Reproduction: Optimal Attention Temperature Improves the Robustness of In-Context Learning under Distribution Shift in High Dimensions

Explore and manage experiment logs with an interactive workspace

Reproduction: Robust and Consistent Ski Rental with Distributional Advice

Explore project logs, code, and traces in an interactive web UI

Reproduction: Dissecting Multimodal In-Context Learning: Modality Asymmetries and Circuit Dynamics in Modern Transformers

Reproduction: Are Two Datasets Close Enough With Statistical Significance? A Kernel Distributional Closeness Testing Approach

Reproduction: Graph Alignment for Benchmarking Graph Neural Networks and Learning Positional Encodings

Explore project code, logs and traces in a web logbook

Reproduction: Prior Diffusiveness and Regret in the Linear-Gaussian Bandit

Browse and manage experiment logs with code, traces, and workspace

Reproduction: The Relative Instability of Model Comparison with Cross-validation

Explore code logs, traces, and workspace in a unified view

Reproduction: High-Probability Convergence Guarantees of Decentralized SGD

Explore and navigate research experiment logbooks

Reproduction: Joint Learning in the Gaussian Single Index Model

Explore code, traces, and workspace in an interactive logbook

Reproduction: Exploration-free Algorithms for Multi-group Mean Estimation

Explore and manage project logs with code, trace, and workspace views

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

Browse experiment logs, traces, and workspace

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

Reproduction: Welfare-Optimal Classification with Accuracy Auctions

Reproduction: A Theoretical Framework for Statistical Evaluability of Generative Models

Explore project logs, traces, and workspace

Reproduction: Belief Propagation Converges to Gaussian Distributions in Sparsely-Connected Factor Graphs

Reproduction: Reward-free Alignment for Conflicting Objectives

Explore project logs, traces, and workspace in one web view

Reproduction: Learning, Solving and Optimizing PDEs with TensorGalerkin: an efficient high-performance Galerkin assembly algorithm

Explore project logs, code, and traces in an interactive web logbook

Reproduction: Improving Sampling for Masked Diffusion Models via Information Gain

Explore experiment logs, traces, and workspace in a web logbook