Za’atari camp is the largest refugee camp in Jordan. It was first established in 2012 to host Syrian refugees. Currently the camp hosts more than 81,000 refugees, with no proper sanitary system which might pose a maj...Za’atari camp is the largest refugee camp in Jordan. It was first established in 2012 to host Syrian refugees. Currently the camp hosts more than 81,000 refugees, with no proper sanitary system which might pose a major threat to surface resources in the area. An investigation was done at Za’atari refugees’ camp to find the impact of refugees settling on surface and groundwater quality. Surface water quality of surface runoff generated from thirty rain fall events were collected during the winter season of 2013/2014 from the major Wadi that passes through the camp and small ponds within the camp after the rainfall event. The collected samples were analyzed for acidity (pH), the electrical connectivity (EC), total dissolved solids (TDS), nutrients (NO3<sup style='margin-left:-7px;'>- and PO4<sup style='margin-left:-7px;'>3-) and selected heavy metals (Mn, Cd, Zn, Pb and Ni) in addition to biological oxygen demand (BOD5), chemical oxygen demand (COD) and intestinal worms (Total Coliform, E. cali). The results showed that there are significant variations in the EC as well as with TDS between the sites due to fluctuating amounts of water used for different activities within the camp as it was highest in the center of the camp where most of the refugees settle decreasing away from the center. The pH values were within the specifications of the World Health Organization and the Jordanian Standards. For nutrients, nitrate concentration was low with high phosphate ions which are most probably from detergents origin.展开更多
The main idea of reinforcement learning is evaluating the chosen action depending on the current reward.According to this concept,many algorithms achieved proper performance on classic Atari 2600 games.The main challe...The main idea of reinforcement learning is evaluating the chosen action depending on the current reward.According to this concept,many algorithms achieved proper performance on classic Atari 2600 games.The main challenge is when the reward is sparse or missing.Such environments are complex exploration environments likeMontezuma’s Revenge,Pitfall,and Private Eye games.Approaches built to deal with such challenges were very demanding.This work introduced a different reward system that enables the simple classical algorithm to learn fast and achieve high performance in hard exploration environments.Moreover,we added some simple enhancements to several hyperparameters,such as the number of actions and the sampling ratio that helped improve performance.We include the extra reward within the human demonstrations.After that,we used Prioritized Double Deep Q-Networks(Prioritized DDQN)to learning from these demonstrations.Our approach enabled the Prioritized DDQNwith a short learning time to finish the first level of Montezuma’s Revenge game and to perform well in both Pitfall and Private Eye.We used the same games to compare our results with several baselines,such as the Rainbow and Deep Q-learning from demonstrations(DQfD)algorithm.The results showed that the new rewards system enabled Prioritized DDQN to out-perform the baselines in the hard exploration games with short learning time.展开更多
强化学习用于序列决策问题上取得的成功越来越受到人们的重视,但是当使用高维状态作为输入时,仍然存在数据效率低下的问题。造成这个问题的原因之一是智能体难以从高维空间提取有效的特征。为了提高数据效率,论文提出一种适用于强化学...强化学习用于序列决策问题上取得的成功越来越受到人们的重视,但是当使用高维状态作为输入时,仍然存在数据效率低下的问题。造成这个问题的原因之一是智能体难以从高维空间提取有效的特征。为了提高数据效率,论文提出一种适用于强化学习任务的数据增强方法cGDA(cGANs-based Data Augment),该方法用条件生成对抗网络(cGANs)对环境的动态特性建模,以当前时刻的状态和动作作为条件生成模型的输入,输出下一时刻的状态作为增强数据。训练过程中使用真实数据和增强数据同时训练智能体,有效地帮助智能体从不同的数据中快速提取到有用的知识。在Atari100K基准上,cGDA在26个离散控制问题环境中与采用数据增强的方法比较,在16个环境中获得了更高的性能;与未采用数据增强的方法比较,在14个环境中获得了更高的性能。展开更多
文摘Za’atari camp is the largest refugee camp in Jordan. It was first established in 2012 to host Syrian refugees. Currently the camp hosts more than 81,000 refugees, with no proper sanitary system which might pose a major threat to surface resources in the area. An investigation was done at Za’atari refugees’ camp to find the impact of refugees settling on surface and groundwater quality. Surface water quality of surface runoff generated from thirty rain fall events were collected during the winter season of 2013/2014 from the major Wadi that passes through the camp and small ponds within the camp after the rainfall event. The collected samples were analyzed for acidity (pH), the electrical connectivity (EC), total dissolved solids (TDS), nutrients (NO3<sup style='margin-left:-7px;'>- and PO4<sup style='margin-left:-7px;'>3-) and selected heavy metals (Mn, Cd, Zn, Pb and Ni) in addition to biological oxygen demand (BOD5), chemical oxygen demand (COD) and intestinal worms (Total Coliform, E. cali). The results showed that there are significant variations in the EC as well as with TDS between the sites due to fluctuating amounts of water used for different activities within the camp as it was highest in the center of the camp where most of the refugees settle decreasing away from the center. The pH values were within the specifications of the World Health Organization and the Jordanian Standards. For nutrients, nitrate concentration was low with high phosphate ions which are most probably from detergents origin.
文摘The main idea of reinforcement learning is evaluating the chosen action depending on the current reward.According to this concept,many algorithms achieved proper performance on classic Atari 2600 games.The main challenge is when the reward is sparse or missing.Such environments are complex exploration environments likeMontezuma’s Revenge,Pitfall,and Private Eye games.Approaches built to deal with such challenges were very demanding.This work introduced a different reward system that enables the simple classical algorithm to learn fast and achieve high performance in hard exploration environments.Moreover,we added some simple enhancements to several hyperparameters,such as the number of actions and the sampling ratio that helped improve performance.We include the extra reward within the human demonstrations.After that,we used Prioritized Double Deep Q-Networks(Prioritized DDQN)to learning from these demonstrations.Our approach enabled the Prioritized DDQNwith a short learning time to finish the first level of Montezuma’s Revenge game and to perform well in both Pitfall and Private Eye.We used the same games to compare our results with several baselines,such as the Rainbow and Deep Q-learning from demonstrations(DQfD)algorithm.The results showed that the new rewards system enabled Prioritized DDQN to out-perform the baselines in the hard exploration games with short learning time.
文摘强化学习用于序列决策问题上取得的成功越来越受到人们的重视,但是当使用高维状态作为输入时,仍然存在数据效率低下的问题。造成这个问题的原因之一是智能体难以从高维空间提取有效的特征。为了提高数据效率,论文提出一种适用于强化学习任务的数据增强方法cGDA(cGANs-based Data Augment),该方法用条件生成对抗网络(cGANs)对环境的动态特性建模,以当前时刻的状态和动作作为条件生成模型的输入,输出下一时刻的状态作为增强数据。训练过程中使用真实数据和增强数据同时训练智能体,有效地帮助智能体从不同的数据中快速提取到有用的知识。在Atari100K基准上,cGDA在26个离散控制问题环境中与采用数据增强的方法比较,在16个环境中获得了更高的性能;与未采用数据增强的方法比较,在14个环境中获得了更高的性能。
