Tianyu Zhan

TSGR: Taobao Search Generative Retrieval

Tianyu Zhan, Gui Ling, Tong Xiong, Kunhai Lin et autres

Generative retrieval (GR) has demonstrated strong promise for industrial e-commerce search by training a single autoregressive model to directly generate the Semantic IDs (SIDs) of target items. However, existing GR systems are primarily optimized for semantic matching and remain insensitive to item …

Multiple Imputation with B-Splines (MI-B) in Clinical Trial Monitoring

Minling Zhang, Jiewei Zeng, Huan Jin, Tianyu Zhan

In clinical trial conduct, it is critical to monitor some important parameters based on accumulating data to ensure quality, inform next-stage planning and protect patients’ safety. The typical approach based on observed outcomes relies on a strong extrapolation assumption, which may be …

us (code pays fourni par la source)

Synergy Area with FDR-controlled Evaluation (SAFE) to robustly assess safety profile in clinical trials

Tianyu Zhan, Yabing Mai, Yihua Gu, Thao Doan et autres

Safety assessment plays a fundamental role in developing a new drug via clinical trials for ethical considerations. Due to complexity, manual review is typically conducted on the totality of data to draw safety conclusions. There are some existing quantitative methods to facilitate …

Synergy Area with FDR-controlled Evaluation (SAFE) to robustly assess safety profile in clinical trials

Tianyu Zhan, Yabing Mai, Yihua Gu, Thao Doan et autres

Safety assessment plays a fundamental role in developing a new drug via clinical trials for ethical considerations. Due to complexity, manual review is typically conducted on the totality of data to draw safety conclusions. There are some existing quantitative methods to facilitate …

us (code pays fourni par la source)

Synergy Area With FDR ‐Controlled Evaluation ( SAFE ) to Robustly Assess Safety Profile in Clinical Trials

Tianyu Zhan, Yabing Mai, Yihua Gu, Thao Doan et autres

Safety assessment plays a fundamental role in developing a new drug via clinical trials for ethical considerations. Due to complexity, manual review is typically conducted on the totality of data to draw safety conclusions. There are some existing quantitative methods to facilitate …

us (code pays fourni par la source)

Semantic Trimming and Auxiliary Multi-step Prediction for Generative Recommendation

Tianyu Zhan, Kairui Fu, Chengfei Lv, Zheqi Lv et autres

Generative Recommendation (GR) has recently transitioned from atomic item-indexing to Semantic ID (SID)-based frameworks to capture intrinsic item relationships and enhance generalization. However, the adoption of high-granularity SIDs leads to two critical challenges: prohibitive training overhead due to sequence expansion and unstable …

Accelerating medicines to patients: A Bayesian quantitative evidence-based decision-making framework with interim monitoring

Yihua Gu, Ziqian Geng, Tianyu Zhan, William R. Henner et autres

us (code pays fourni par la source)

Lutikizumab in Adults With Moderate to Severe Hidradenitis Suppurativa After Anti-TNF Therapy Failure

Alexa B. Kimball, Konrad Teodor Sawicki, Lindsay S. Ackerman, Hermenio Lima et autres

Importance: Hidradenitis suppurativa (HS) is a debilitating inflammatory skin disease with limited therapeutic options. Objective: To assess the efficacy and safety of lutikizumab, a dual-variable domain interleukin 1α/1β antagonist, for adult participants with moderate to severe HS who experienced anti-tumor necrosis factor …

us, ca, gr, de (code pays fourni par la source)

Some Future Works

Tianyu Zhan

In this chapter, we discuss several extensions or future works based on the previously introduced DNN -assisted methods. For example, the two-group comparison with the parametric assumption in Chapter 3 can be generalized to nonparametric hypothesis testing and multiple comparisons. For the …

Introduction to Deep Neural Networks (DNNs)

Tianyu Zhan

This chapter provides a high-level introduction to this textbook on assisting several statistical methods with deep learning. Specifically, deep learning is a prominent machine learning method, which is a subfield within artificial intelligence. As compared to some traditional machine learning methods, deep …

Interpretable Models in Regression Analysis

Tianyu Zhan

An interpretable model is preferred in practice due to several advantages, such as being robust to outliers, offering a better understanding of the underlying scientific question, and facilitating external communications. However, there is no standard approach to evaluate interpretation, and even its …

Deep-Learning-Assisted Statistical Methods with Examples in R

Tianyu Zhan

This book explores how deep learning enhances statistical methods for hypothesis testing, point estimation, optimization, interpretation, and other aspects. It uniquely demonstrates leveraging deep learning to improve traditional statistical approaches, showcasing their superior performance in practical applications. Each topic includes essential background, …