Cheng Zhen

Minimal Data Cleaning for Model Training by MinPrep

Cheng Zhen, Prayoga, Nischal Aryal, Arash Termehchy et autres

Real-world datasets often contain dirty data, such as missing values, outliers, and inconsistencies. To train accurate machine learning (ML) models over these datasets, users typically spend a substantial amount of time and resources determining proper replacement values for dirty entries in training …

us (code pays fourni par la source)

CustomListener: Text-Guided Responsive Interaction for User-Friendly Listening Head Generation

Xi Liu, Ying Guo, Cheng Zhen, Tong Li et autres

Listening head generation aims to synthesize a non-verbal responsive listener head by modeling the correlation between the speaker and the listener in dynamic conversion. The applications of listener agent generation in virtual interaction have promoted many works achieving diverse and fine-grained motion …

cn (code pays fourni par la source)

Certain and Approximately Certain Models for Statistical Learning

Cheng Zhen, Nischal Aryal, Arash Termehchy, Amandeep Singh Chabada

Real-world data is often incomplete and contains missing values. To train accurate models over real-world datasets, users need to spend a substantial amount of time and resources imputing and finding proper values for missing data items. In this paper, we demonstrate that …

us (code pays fourni par la source)

CustomListener: Text-guided Responsive Interaction for User-friendly Listening Head Generation

Xi Liu, Ying Guo, Cheng Zhen, Tong Li et autres

Listening head generation aims to synthesize a non-verbal responsive listener head by modeling the correlation between the speaker and the listener in dynamic conversion.The applications of listener agent generation in virtual interaction have promoted many works achieving the diverse and fine-grained motion …

When Can We Ignore Missing Data in Model Training?

Cheng Zhen, Amandeep Singh Chabada, Arash Termehchy

Imputing missing data is typically expensive, and as a result, people seek to avoid it when possible. To address this issue, we introduce a method that determines when data cleaning is unnecessary for machine learning (ML). If a model can minimize the …

us (code pays fourni par la source)

Feature aided Gaussian mixture probability hypothesis density filter with modified 2D assignment

Cheng Zhen, Wen Shu-liang

In order to track multiple targets with time-varying number of targets, the paper proposed a new feature aided Gaussian mixture probability hypothesis density (FA-GM-PHD) filter, and adopts a modified 2-D assignment algorithm to carry out the data association and manage the tracks …

cn (code pays fourni par la source)