The Experts below are selected from a list of 102627 Experts worldwide ranked by ideXlab platform
Yanmin Zhu - One of the best experts on this subject based on the ideXlab platform.
-
Targeted Knowledge transfer for learning traffic signal plans
Pacific-Asia Conference on Knowledge Discovery and Data Mining, 2019Co-Authors: Guanjie Zheng, Yanmin ZhuAbstract:Traffic signal control in cities today is not well optimized according to the feedback received from the real world. And such an inefficiency in traffic signal control results in people’s waste of time in commuting, road rage in the traffic jam, and high cost for city operation. Recently, deep reinforcement learning (DRL) approaches shed lights to better optimize traffic signal plans according to the feedback received from the environment. Most of these methods are evaluated in a simulated environment, but can not be applied to intersections in the real world directly, as the training of DRL relies on a great amount of samples and takes a long time to converge. In this paper, we propose a batch learning framework where the Targeted transfer reinforcement learning (TTRL-B) is introduced to speed up learning. Specifically, a separate unsupervised method is designed to measure the similarities of traffic conditions to select the suitable source intersection for transfer. The proposed framework allows batch learning and this is the first work to consider the impact of slow learning in RL on real-world applications. Experiments on real traffic data demonstrate that our model accelerates learning with good performance.
-
PAKDD (2) - Targeted Knowledge Transfer for Learning Traffic Signal Plans
Advances in Knowledge Discovery and Data Mining, 2019Co-Authors: Guanjie Zheng, Yanmin ZhuAbstract:Traffic signal control in cities today is not well optimized according to the feedback received from the real world. And such an inefficiency in traffic signal control results in people’s waste of time in commuting, road rage in the traffic jam, and high cost for city operation. Recently, deep reinforcement learning (DRL) approaches shed lights to better optimize traffic signal plans according to the feedback received from the environment. Most of these methods are evaluated in a simulated environment, but can not be applied to intersections in the real world directly, as the training of DRL relies on a great amount of samples and takes a long time to converge. In this paper, we propose a batch learning framework where the Targeted transfer reinforcement learning (TTRL-B) is introduced to speed up learning. Specifically, a separate unsupervised method is designed to measure the similarities of traffic conditions to select the suitable source intersection for transfer. The proposed framework allows batch learning and this is the first work to consider the impact of slow learning in RL on real-world applications. Experiments on real traffic data demonstrate that our model accelerates learning with good performance.
Guanjie Zheng - One of the best experts on this subject based on the ideXlab platform.
-
Targeted Knowledge transfer for learning traffic signal plans
Pacific-Asia Conference on Knowledge Discovery and Data Mining, 2019Co-Authors: Guanjie Zheng, Yanmin ZhuAbstract:Traffic signal control in cities today is not well optimized according to the feedback received from the real world. And such an inefficiency in traffic signal control results in people’s waste of time in commuting, road rage in the traffic jam, and high cost for city operation. Recently, deep reinforcement learning (DRL) approaches shed lights to better optimize traffic signal plans according to the feedback received from the environment. Most of these methods are evaluated in a simulated environment, but can not be applied to intersections in the real world directly, as the training of DRL relies on a great amount of samples and takes a long time to converge. In this paper, we propose a batch learning framework where the Targeted transfer reinforcement learning (TTRL-B) is introduced to speed up learning. Specifically, a separate unsupervised method is designed to measure the similarities of traffic conditions to select the suitable source intersection for transfer. The proposed framework allows batch learning and this is the first work to consider the impact of slow learning in RL on real-world applications. Experiments on real traffic data demonstrate that our model accelerates learning with good performance.
-
PAKDD (2) - Targeted Knowledge Transfer for Learning Traffic Signal Plans
Advances in Knowledge Discovery and Data Mining, 2019Co-Authors: Guanjie Zheng, Yanmin ZhuAbstract:Traffic signal control in cities today is not well optimized according to the feedback received from the real world. And such an inefficiency in traffic signal control results in people’s waste of time in commuting, road rage in the traffic jam, and high cost for city operation. Recently, deep reinforcement learning (DRL) approaches shed lights to better optimize traffic signal plans according to the feedback received from the environment. Most of these methods are evaluated in a simulated environment, but can not be applied to intersections in the real world directly, as the training of DRL relies on a great amount of samples and takes a long time to converge. In this paper, we propose a batch learning framework where the Targeted transfer reinforcement learning (TTRL-B) is introduced to speed up learning. Specifically, a separate unsupervised method is designed to measure the similarities of traffic conditions to select the suitable source intersection for transfer. The proposed framework allows batch learning and this is the first work to consider the impact of slow learning in RL on real-world applications. Experiments on real traffic data demonstrate that our model accelerates learning with good performance.
Julie Rochester - One of the best experts on this subject based on the ideXlab platform.
-
an illustration of delivering evidence based practice through staff training multi dimensional process outcome and organizational evaluation
Behavioural and Cognitive Psychotherapy, 2003Co-Authors: Derek Milne, Katy Woodward, Seirian Hanner, John Iceton, Ann Fitzsimmons, Julie RochesterAbstract:Developing staff competence in evidence-based interventions for people with a severe mental illnesses is now a government priority. However, evidence concerning the nature and effectiveness of the training programmes that are designed to develop competence is decidedly limited. The purpose of the present study was, therefore, to evaluate systematically a manualized and evidence-based staff training programme. This programme was delivered over a 17-day period, in a modular and experiential workshop format. Staff (mostly mental health nurses) were allocated to either an experimental group (N = 18) who received the training or to a waiting-list, control group condition (N = 7). Multi-dimensional process and outcome evaluations indicated that the staff training was delivered competently, followed an experiential learning approach, and led to significant improvement in the participants' Targeted Knowledge and skills. The staff were also satisfied with the acceptability and delivery of the training. Generalization to the participants' clients was reported to be very good, which was congruent with evidence of a supportive work environment, the organizational evaluation aspect of this study. While these learning outcomes are typical of such training, the degree of generalization is exceptional. Possible reasons for this favourable outcome are discussed, and conclusions are drawn for developing staff competence in ways that maximize the delivery of evidence-based practice.
Derek Milne - One of the best experts on this subject based on the ideXlab platform.
-
an illustration of delivering evidence based practice through staff training multi dimensional process outcome and organizational evaluation
Behavioural and Cognitive Psychotherapy, 2003Co-Authors: Derek Milne, Katy Woodward, Seirian Hanner, John Iceton, Ann Fitzsimmons, Julie RochesterAbstract:Developing staff competence in evidence-based interventions for people with a severe mental illnesses is now a government priority. However, evidence concerning the nature and effectiveness of the training programmes that are designed to develop competence is decidedly limited. The purpose of the present study was, therefore, to evaluate systematically a manualized and evidence-based staff training programme. This programme was delivered over a 17-day period, in a modular and experiential workshop format. Staff (mostly mental health nurses) were allocated to either an experimental group (N = 18) who received the training or to a waiting-list, control group condition (N = 7). Multi-dimensional process and outcome evaluations indicated that the staff training was delivered competently, followed an experiential learning approach, and led to significant improvement in the participants' Targeted Knowledge and skills. The staff were also satisfied with the acceptability and delivery of the training. Generalization to the participants' clients was reported to be very good, which was congruent with evidence of a supportive work environment, the organizational evaluation aspect of this study. While these learning outcomes are typical of such training, the degree of generalization is exceptional. Possible reasons for this favourable outcome are discussed, and conclusions are drawn for developing staff competence in ways that maximize the delivery of evidence-based practice.
Elena Giglia - One of the best experts on this subject based on the ideXlab platform.
-
Targeted Knowledge interaction and rich user experience towards a scholarly communication that lets
International Conference on Electronic Publishing, 2009Co-Authors: G P Pescarmona, Elena GigliaAbstract:All living systems share many properties including hardly predictable behaviours, due to the differences between individuals and the chaos in natural environments. The reductionist approach to the interpretation of these phenomena suffers from the oversimplification of the factors involved in the quest of universal “scientific” explanations. The validation of scientific paradigms is based on the consensus of leading groups that decide what is true and what is not. That means that all events - not only conflicting opinions but also conflicting raw data - not fitting with the scientific official truth were never published, and that supported indirectly the correctness of the experts’ choice. With the advent of Web 2.0 and the freedom of publishing, the number of these not fitting events has dramatically increased. Yesterday, data were supplied to the reader with the interpretation. Now the reader has to afford in each field a huge amount of data and opinions. Extracting from the garbage the information you need requires a strategy. Strategy is a science by itself. In the specific case of Knowledge the first step is to define Knowledge. The aim of life sciences, medicine, social sciences is to modify the reality when it is no longer sustainable, whatever it could mean in every single situation. I have to know how my system works to
-
ELPUB - Targeted Knowledge: INTERACTION AND RICH USER EXPERIENCE TOWARDS A SCHOLARLY COMMUNICATION THAT "LETS"
2009Co-Authors: G P Pescarmona, Elena GigliaAbstract:All living systems share many properties including hardly predictable behaviours, due to the differences between individuals and the chaos in natural environments. The reductionist approach to the interpretation of these phenomena suffers from the oversimplification of the factors involved in the quest of universal “scientific” explanations. The validation of scientific paradigms is based on the consensus of leading groups that decide what is true and what is not. That means that all events - not only conflicting opinions but also conflicting raw data - not fitting with the scientific official truth were never published, and that supported indirectly the correctness of the experts’ choice. With the advent of Web 2.0 and the freedom of publishing, the number of these not fitting events has dramatically increased. Yesterday, data were supplied to the reader with the interpretation. Now the reader has to afford in each field a huge amount of data and opinions. Extracting from the garbage the information you need requires a strategy. Strategy is a science by itself. In the specific case of Knowledge the first step is to define Knowledge. The aim of life sciences, medicine, social sciences is to modify the reality when it is no longer sustainable, whatever it could mean in every single situation. I have to know how my system works to
