Search Results - "Li, Hongkang"

Colorimetric Aerogel Gas Sensor with High Sensitivity and Stability by Xia, Xiaoli, Wu, Ruonan, Zhang, Lei, Chen, Xiangyu, Yan, Yanling, Yin, Jikun, Ren, Jin, Li, Hongkang, Yin, Jinzhong, Xue, Zhenjie, Yi, Lanhua, Wang, Tie

“…The detection of formic acid vapor in the usage environment is extremely important for human health and safety. The utilization of metal–organic frameworks…”
Get full text

How Does Promoting the Minority Fraction Affect Generalization? A Theoretical Study of One-Hidden-Layer Neural Network on Group Imbalance by Li, Hongkang, Zhang, Shuai, Zhang, Yihua, Wang, Meng, Liu, Sijia, Chen, Pin-Yu

“…Group imbalance has been a known problem in empirical risk minimization (ERM), where the achieved high average accuracy is accompanied by low accuracy in a…”
Get full text

How Can Personalized Context Help? Exploring Joint Retrieval of Passage and Personalized Context by Wan, Hui, Li, Hongkang, Lu, Songtao, Cui, Xiaodong, Danilevsky, Marina

“…The integration of external personalized context information into document-grounded conversational systems has significant potential business value, but has…”
Get full text

Learning on Transformers is Provable Low-Rank and Sparse: A One-layer Analysis by Li, Hongkang, Wang, Meng, Zhang, Shuai, Liu, Sijia, Chen, Pin-Yu

“…Efficient training and inference algorithms, such as low-rank adaption and model pruning, have shown impressive performance for learning Transformer-based…”
Get full text

Learning and generalization of one-hidden-layer neural networks, going beyond standard Gaussian data by Li, Hongkang, Zhang, Shuai, Wang, Meng

“…This paper analyzes the convergence and generalization of training a one-hidden-layer neural network when the input features follow the Gaussian mixture model…”
Get full text

Learning and generalization of one-hidden-layer neural networks, going beyond standard Gaussian data by Li, Hongkang, Zhang, Shuai, Wang, Meng

“…This paper analyzes the convergence and generalization of training a one-hidden-layer neural network when the input features follow the Gaussian mixture model…”
Get full text

How Can Context Help? Exploring Joint Retrieval of Passage and Personalized Context by Wan, Hui, Li, Hongkang, Lu, Songtao, Cui, Xiaodong, Danilevsky, Marina

“…The integration of external personalized context information into document-grounded conversational systems has significant potential business value, but has…”
Get full text

Enhancing Graph Transformers with Hierarchical Distance Structural Encoding by Luo, Yuankai, Li, Hongkang, Shi, Lei, Wu, Xiao-Ming

“…Graph transformers need strong inductive biases to derive meaningful attention scores. Yet, current methods often fall short in capturing longer ranges,…”
Get full text

A Theoretical Understanding of Shallow Vision Transformers: Learning, Generalization, and Sample Complexity by Li, Hongkang, Wang, Meng, Liu, Sijia, Chen, Pin-yu

“…ICLR 2023 Vision Transformers (ViTs) with self-attention modules have recently achieved great empirical success in many vision tasks. Due to non-convex…”
Get full text

Training Nonlinear Transformers for Chain-of-Thought Inference: A Theoretical Generalization Analysis by Li, Hongkang, Wang, Meng, Lu, Songtao, Cui, Xiaodong, Chen, Pin-Yu

“…Chain-of-Thought (CoT) is an efficient prompting method that enables the reasoning ability of large language models by augmenting the query using multiple…”
Get full text

Learning on Transformers is Provable Low-Rank and Sparse: A One-layer Analysis by Li, Hongkang, Wang, Meng, Zhang, Shuai, Liu, Sijia, Chen, Pin-Yu

“…Efficient training and inference algorithms, such as low-rank adaption and model pruning, have shown impressive performance for learning Transformer-based…”
Get full text

What Improves the Generalization of Graph Transformers? A Theoretical Dive into the Self-attention and Positional Encoding by Li, Hongkang, Wang, Meng, Ma, Tengfei, Liu, Sijia, Zhang, Zaixi, Chen, Pin-Yu

“…Graph Transformers, which incorporate self-attention and positional encoding, have recently emerged as a powerful architecture for various graph learning…”
Get full text

Node Identifiers: Compact, Discrete Representations for Efficient Graph Learning by Luo, Yuankai, Li, Hongkang, Liu, Qijiong, Shi, Lei, Wu, Xiao-Ming

“…We present a novel end-to-end framework that generates highly compact (typically 6-15 dimensions), discrete (int4 type), and interpretable node…”
Get full text

How does promoting the minority fraction affect generalization? A theoretical study of the one-hidden-layer neural network on group imbalance by Li, Hongkang, Zhang, Shuai, Zhang, Yihua, Wang, Meng, Liu, Sijia, Chen, Pin-Yu

“…Group imbalance has been a known problem in empirical risk minimization (ERM), where the achieved high average accuracy is accompanied by low accuracy in a…”
Get full text

How Do Nonlinear Transformers Learn and Generalize in In-Context Learning? by Li, Hongkang, Wang, Meng, Lu, Songtao, Cui, Xiaodong, Chen, Pin-Yu

“…Transformer-based large language models have displayed impressive in-context learning capabilities, where a pre-trained model can handle new tasks without…”
Get full text

On the Convergence and Sample Complexity Analysis of Deep Q-Networks with $\epsilon$-Greedy Exploration by Zhang, Shuai, Li, Hongkang, Wang, Meng, Liu, Miao, Chen, Pin-Yu, Lu, Songtao, Liu, Sijia, Murugesan, Keerthiram, Chaudhury, Subhajit

“…Neurips 2023 This paper provides a theoretical understanding of Deep Q-Network (DQN) with the $\varepsilon$-greedy exploration in deep reinforcement learning…”
Get full text

Generalization Guarantee of Training Graph Convolutional Networks with Graph Topology Sampling by Li, Hongkang, Wang, Meng, Liu, Sijia, Chen, Pin-Yu, Xiong, Jinjun

“…ICML 2022 Graph convolutional networks (GCNs) have recently achieved great empirical success in learning graph-structured data. To address its scalability…”
Get full text