Few multi-agent reinforcement learning (MARL) researches on Google research football (GRF) focus on the 11-vs-11 multi-agent full-game scenario and to the best of our knowledge, no open benchmark on this scenario has ...Few multi-agent reinforcement learning (MARL) researches on Google research football (GRF) focus on the 11-vs-11 multi-agent full-game scenario and to the best of our knowledge, no open benchmark on this scenario has been released to the public. In this work, we fill the gap by providing a population-based MARL training pipeline and hyperparameter settings on multi-agent football scenario that outperforms the bot with difficulty 1.0 from scratch within 2 million steps. Our experiments serve as a reference for the expected performance of independent proximal policy optimization (IPPO), a state-of-the-art multi-agent reinforcement learning algorithm where each agent tries to maximize its own policy independently across various training configurations. Meanwhile, we release our training framework Light-MALib which extends the MALib codebase by distributed and asynchronous implementation with additional analytical tools for football games. Finally, we provide guidance for building strong football AI with population-based training and release diverse pretrained policies for benchmarking. The goal is to provide the community with a head start for whoever experiment their works on GRF and a simple-to-use population-based training framework for further improving their agents through self-play. The implementation is available at https://github.com/Shanghai-Digital-Brain-Laboratory/DB-Football.展开更多
LED has shown great advantages in poultry husbandry.This study focused on the behavioral preferences and production performance of chicken broilers reared under unevenly distributed yellow LED light.Four pens were div...LED has shown great advantages in poultry husbandry.This study focused on the behavioral preferences and production performance of chicken broilers reared under unevenly distributed yellow LED light.Four pens were divided into two groups adopting respective maximum light intensities(MLIs,60 lx and 30 lx).Because of different distances from the installation position of the LED pipe,each pen was distributed with unevenly distributed light.Each pen consisted of four subzones indicated by their light intensities-high intensity(HI),medium intensity(MI),low intensity(LI)and weak intensity(WI).Four subzones were the same size and provided with a feeder and a drinker,respectively.The broilers moved freely across the four subzones.No significant differences in body weight(BW),uniformity of final BW and feed conversion ratio(FCR)were observed between the two experimental groups.However,greater feed intake and water intake were found in HI than those in other subzones.The drinking preference changed with age for four subzones and was more likely to appear at the later stage,despite substantial fluctuations within the replicates.The feeding preference was more constant than the drinking preference and appeared mainly at the early and middle stages of this study.These findings could provide implications for broiler production reared under unevenly distributed LED light condition.展开更多
It is an important task to improve performance for sparse matrix vector multiplication (SpMV), and it is a difficult task because of its irregular memory access. Gen- eral purpose GPU (GPGPU) provides high computi...It is an important task to improve performance for sparse matrix vector multiplication (SpMV), and it is a difficult task because of its irregular memory access. Gen- eral purpose GPU (GPGPU) provides high computing abil- ity and substantial bandwidth that cannot be fully exploited by SpMV due to its irregularity. In this paper, we propose two novel methods to optimize the memory bandwidth for SpMV on GPGPU. First, a new storage format is proposed to exploit memory bandwidth of GPU architecture more effi- ciently. The new storage format can ensure that there are as many non-zeros as possible in the format which is suitable to exploit the memory bandwidth of the GPU. Second, we pro- pose a cache blocking method to improve the performance of SpMV on GPU architecture. The sparse matrix is partitioned into sub-blocks that are stored in CSR format. With the block- ing method, the corresponding part of vector x can be reused in the GPU cache, so the time to access the global memory for vector x is reduced heavily. Experiments are carried out on three GPU platforms, GeForce 9800 GX2, GeForce GTX 480, and Tesla K40. Experimental results show that both new methods can efficiently improve the utilization of GPU mem- ory bandwidth and the performance of the GPU.展开更多
基金supported by the National Natural Science Foundation of China(No.62206289).
文摘Few multi-agent reinforcement learning (MARL) researches on Google research football (GRF) focus on the 11-vs-11 multi-agent full-game scenario and to the best of our knowledge, no open benchmark on this scenario has been released to the public. In this work, we fill the gap by providing a population-based MARL training pipeline and hyperparameter settings on multi-agent football scenario that outperforms the bot with difficulty 1.0 from scratch within 2 million steps. Our experiments serve as a reference for the expected performance of independent proximal policy optimization (IPPO), a state-of-the-art multi-agent reinforcement learning algorithm where each agent tries to maximize its own policy independently across various training configurations. Meanwhile, we release our training framework Light-MALib which extends the MALib codebase by distributed and asynchronous implementation with additional analytical tools for football games. Finally, we provide guidance for building strong football AI with population-based training and release diverse pretrained policies for benchmarking. The goal is to provide the community with a head start for whoever experiment their works on GRF and a simple-to-use population-based training framework for further improving their agents through self-play. The implementation is available at https://github.com/Shanghai-Digital-Brain-Laboratory/DB-Football.
基金We acknowledge the support from National Key R&D Program of China(Grant No.2017YFB0404000)China Postdoctoral Science Foundation(2018M632470)Postdoctoral Science Foundation of Zhejiang Province.
文摘LED has shown great advantages in poultry husbandry.This study focused on the behavioral preferences and production performance of chicken broilers reared under unevenly distributed yellow LED light.Four pens were divided into two groups adopting respective maximum light intensities(MLIs,60 lx and 30 lx).Because of different distances from the installation position of the LED pipe,each pen was distributed with unevenly distributed light.Each pen consisted of four subzones indicated by their light intensities-high intensity(HI),medium intensity(MI),low intensity(LI)and weak intensity(WI).Four subzones were the same size and provided with a feeder and a drinker,respectively.The broilers moved freely across the four subzones.No significant differences in body weight(BW),uniformity of final BW and feed conversion ratio(FCR)were observed between the two experimental groups.However,greater feed intake and water intake were found in HI than those in other subzones.The drinking preference changed with age for four subzones and was more likely to appear at the later stage,despite substantial fluctuations within the replicates.The feeding preference was more constant than the drinking preference and appeared mainly at the early and middle stages of this study.These findings could provide implications for broiler production reared under unevenly distributed LED light condition.
文摘It is an important task to improve performance for sparse matrix vector multiplication (SpMV), and it is a difficult task because of its irregular memory access. Gen- eral purpose GPU (GPGPU) provides high computing abil- ity and substantial bandwidth that cannot be fully exploited by SpMV due to its irregularity. In this paper, we propose two novel methods to optimize the memory bandwidth for SpMV on GPGPU. First, a new storage format is proposed to exploit memory bandwidth of GPU architecture more effi- ciently. The new storage format can ensure that there are as many non-zeros as possible in the format which is suitable to exploit the memory bandwidth of the GPU. Second, we pro- pose a cache blocking method to improve the performance of SpMV on GPU architecture. The sparse matrix is partitioned into sub-blocks that are stored in CSR format. With the block- ing method, the corresponding part of vector x can be reused in the GPU cache, so the time to access the global memory for vector x is reduced heavily. Experiments are carried out on three GPU platforms, GeForce 9800 GX2, GeForce GTX 480, and Tesla K40. Experimental results show that both new methods can efficiently improve the utilization of GPU mem- ory bandwidth and the performance of the GPU.