Search Results - "Huang, ShengYi"

An Empirical Investigation of Early Stopping Optimizations in Proximal Policy Optimization by Dossa, Rousslan Fernand Julien, Huang, Shengyi, Ontanon, Santiago, Matsubara, Takashi

“…Code-level optimizations, which are low-level optimization techniques used in the implementation of algorithms, have generally been considered as tangential…”
Get full text

Prognostic impact of preoperative prognostic nutritional index in resected advanced gastric cancer: A multicenter propensity score analysis by Luo, Zeyu, Zhou, Lin, Balde, Alpha I., Li, Zhou, He, Linyun, ZhenWei, Cai, Zou, ZeNan, Huang, ShengYi, Han, Shuai, Wei Zhou, Min, Zhang, Gang Qing, Cai, Zhai

“…Advanced gastric cancer (AGC) causes debilitating malnutrition and leads to deterioration of the immune response. However, the concept of the prognostic…”
Get full text

Comparison of the outcomes of cytoreductive surgery versus surgery plus hyperthermic intraperitoneal chemotherapy for peritoneal carcinomatosis: a propensity score matching analysis by Li, Zhou, Redondo Ntutumu, Juan de Dios, Huang, Shengyi, Cai, Zhai, Han, Shuai, Balde, A. I., Luo, Zeyu, Fang, Suzhen

“…Background Cytoreductive surgery (CRS) and hyperthermic intraperitoneal chemotherapy (HIPEC) are effective treatment options for selected patients with…”
Get full text

Variations in plant–microbe–soil C:N:P stoichiometry along a 900-year age gradient in Torreya grandis ‘Merrillii’ plantations in Southeast China by He, Sijia, Huang, Juying, Huang, Shengyi, Fang, Zhao, Zhang, Shuoxin, Zhou, Zhichun, Wang, Bin

“…Researches on the ecological stoichiometry of forest vegetation at different growth stages under long-term human management activities and its driving factors…”
Get full text

A conceptual study on the formulation of a permeable reactive pavement with activated carbon additives for controlling the fate of non-point source environmental organic contaminants by Huang, Shengyi, Liang, Chenju

“…To take advantage of the road pavement network where non-point source (NPS) pollution such as benzene, toluene, ethyl-benzene, and xylene (BTEX) from vehicle…”
Get full text

A column study of persulfate chemical oxidative regeneration of toluene gas saturated activated carbon by Jatta, Simon, Huang, Shengyi, Liang, Chenju

“…[Display omitted] •Thermal activated persulfate (TAP) for regenerating toluene gas saturated AC was studied.•The TAP regenerated ACs retained >90% of…”
Get full text

Reproducible and Efficient Deep Reinforcement Learning by Huang, Shengyi

“…Deep reinforcement learning (DRL), a paradigm by which agents learn how to do tasks through trial and error, has achieved great success in many domains…”
Get full text

Factors driving the assembly of prokaryotic communities in bulk soil and rhizosphere of Torreya grandis along a 900-year age gradient by Wang, Bin, Huang, Shengyi, Li, Zhengcai, Zhou, Zhichun, Huang, Juying, Yu, Hailong, Peng, Tong, Song, Yanfang, Na, Xiaofan

“…Excessive nutrient inputs imperil the stability of forest ecosystems via modifying the interactions among soil properties, microbes, and plants, particularly…”
Get full text

Persulfate Chemical Functionalization of Carbon Nanotubes and Associated Adsorption Behavior in Aqueous Phase by Huang, Shengyi, Liang, Chenju, Chen, Yan-Jyun

“…The chemical functionalization of carbon nanotubes (CNTs) using sodium persulfate (SPS) oxidation was designed to improve their dispersion stability in water…”
Get full text

Differences in the dielectric properties of various benign and malignant thyroid nodules by Huang, Shengyi, Cai, Weizhen, Han, Shuai, lin, Yu, Wang, Yu, Chen, Fei, Shao, Guoli, Liu, Yonghong, Yu, Xuefei, Cai, Zhai, Zou, Zenan, Yao, Shun, Wang, Qiaohui, Li, Zhou

“…Purpose This experiment was conducted to investigate the dielectric properties of different types of thyroid nodules. Our goal was to find a simple and fast…”
Get full text

B/Al Codoped/Coated Ultra-High Nickel Cobalt-Free Material with Excellent High Voltage/Rate Cycle Stability by Zhang, Liang, Huang, Jinfu, Tang, Hongyu, Huang, Shengyi, Tang, Yang, Ma, Jianyao, Yang, Jianwen, Huang, Bin, Li, Yanwei, Xiao, Shunhua

“…Ultrahigh nickel cobalt-free cathode materials have high energy density and are very promising materials for application in lithium-ion batteries. However,…”
Get full text

Enhanced stratospheric intrusion at Lulin Mountain, Taiwan inferred from beryllium-7 activity by Huang, Shengyi, Huang, Pin-Ru, Newman, Sally, Li, King-Fai, Lin, Yu-Chi, Huh, Chih-An, Lin, Neng-Huei, Hsu, Shih-Chieh, Liang, Mao-Chang

“…Beryllium-7 (7Be), produced by the interaction of cosmic radiation with atoms and molecules primarily in the upper troposphere and lower stratosphere, provides…”
Get full text

A Closer Look at Invalid Action Masking in Policy Gradient Algorithms by Huang, Shengyi, Ontañón, Santiago

“…FLAIRS. Vol. 35 (2022) In recent years, Deep Reinforcement Learning (DRL) algorithms have achieved state-of-the-art performance in many challenging strategy…”
Get full text

Action Guidance: Getting the Best of Sparse Rewards and Shaped Rewards for Real-time Strategy Games by Huang, Shengyi, Ontañón, Santiago

“…Training agents using Reinforcement Learning in games with sparse rewards is a challenging problem, since large amounts of exploration are required to retrieve…”
Get full text

Comparing Observation and Action Representations for Deep Reinforcement Learning in $\mu$RTS by Huang, Shengyi, Ontañón, Santiago

“…This paper presents a preliminary study comparing different observation and action space representations for Deep Reinforcement Learning (DRL) in the context…”
Get full text

Asynchronous RLHF: Faster and More Efficient Off-Policy RL for Language Models by Noukhovitch, Michael, Huang, Shengyi, Xhonneux, Sophie, Hosseini, Arian, Agarwal, Rishabh, Courville, Aaron

“…The dominant paradigm for RLHF is online and on-policy RL: synchronously generating from the large language model (LLM) policy, labelling with a reward model,…”
Get full text

The N+ Implementation Details of RLHF with PPO: A Case Study on TL;DR Summarization by Huang, Shengyi, Noukhovitch, Michael, Hosseini, Arian, Rasul, Kashif, Wang, Weixun, Tunstall, Lewis

“…This work is the first to openly reproduce the Reinforcement Learning from Human Feedback (RLHF) scaling behaviors reported in OpenAI's seminal TL;DR…”
Get full text

Griddly: A platform for AI research in games by Bamford, Chris, Huang, Shengyi, Lucas, Simon

“…In recent years, there have been immense breakthroughs in Game AI research, particularly with Reinforcement Learning (RL). Despite their success, the…”
Get full text

Reward Scale Robustness for Proximal Policy Optimization via DreamerV3 Tricks by Sullivan, Ryan, Kumar, Akarsh, Huang, Shengyi, Dickerson, John P, Suarez, Joseph

“…Most reinforcement learning methods rely heavily on dense, well-normalized environment rewards. DreamerV3 recently introduced a model-based method with a…”
Get full text

Cleanba: A Reproducible and Efficient Distributed Reinforcement Learning Platform by Huang, Shengyi, Weng, Jiayi, Charakorn, Rujikorn, Lin, Min, Xu, Zhongwen, Ontañón, Santiago

“…Distributed Deep Reinforcement Learning (DRL) aims to leverage more computational resources to train autonomous agents with less training time. Despite recent…”
Get full text