27 research outputs found
Motor kontrol ve beynin bilişsel karar verme mekanizmalarını analiz etmek üzere tersine pekiştirmeli öğrenme ile keşifler
No full textReinforcement Learning is a framework for generating optimal policies given a task and a reward/punishment structure. Likewise, Inverse Reinforcement Learning, as the name suggests, is used for recovering the reasoning behind an optimal policy based on demonstrations from an expert. We set out to explore whether recent Reinforcement Learning and Inverse Reinforcement Learning methods can serve as a computational tool for investigating optimality principles of motor control and cognitive decision-making mechanisms of the brain. For this purpose, we have targeted several different tasks involved with different parts of the sensorimotor learning mechanism of the brain. We aim to recover the optimality principles employed by the brain for various control and decision-making tasks. If this is achieved, we can analyze, understand, mimic and improve demonstrated behavior with less bias, which we hope is a step forward in understanding the process of learning in both human-based and artificial systems. For the scope of this thesis, we have evaluated two tasks. The first task was investigating the applicability of perceptual development for Reinforcement Learning. For this task, we have proposed a perceptual development based learning regime for a Reinforcement Learning agent, and the results obtained suggest that a suitable perceptual development regime may improve the learning progress and yield better-performing agents. The second task was to predict reward function parameters of a provided trajectory in a standing up under perturbation scenario. For this task, we have proposed two different Inverse Reinforcement Learning approaches. Our results indicate that we were able to infer valid reward parameters on synthetic data.Pekiştirmeli öğrenme, farklı ortamlarda, verilen ödül ceza yapısına göre en uygun politikaları bulma sistemidir. Benzer şekilde, Tersine Pekiştirmeli Öğrenme de, adından anlaşılabileceği gibi, bir uzmandan alınan en uygun politikanın arkasındaki sebepleri bulmak için kullanılır. Bu araştırmada, güncel Pekiştirmeli Öğrenme ve Tersine Pekiştirmeli Öğrenme metotlarının, beynin motor kontrol ve bilişsel karar alma mekanizmalarının arkasındaki eniyileme prensiplerini modelleyen araçlar olarak kullanılabilmesini keşfetmeyi amaçlıyoruz. Bu amaç için, beynin farklı duyusal motor özelliklerini hedefleyen farklı görevleri hedefledik. Niyetimiz, beyin tarafından farklı alanlar için oluşturulan en iyileme kriterlerini keşfedebilmek. Bu başarılabildiği takdirde, varolan veya yeni bir metot ile, insan davranışlarını daha düşük bir yanlılık ile analiz edebilir, anlayabilir ve taklit edebiliriz. Bu tezin kapsamı doğrultusunda, iki tane görevi hedefledik. İlk görev, algısal gelişimin Pekiştirmeli Öğrenme'ye uygulanabilirliğinin araştırılmasıdır. Bu görev için, bir Pekiştirmeli Öğrenme ajanı, kendi önerdiğimiz bir algısal gelişim tabanlı gelişimsel rejim ile eğittik. Sonuçlarımız, uygun bir algısal gelişim rejiminin, Pekiştirmeli Öğrenme'nin öğrenme ilerlemesini geliştirebileceğini ve daha iyi ajanlar üretebileceğini önerdi. İkinci görev ise, Tersine Pekiştirmeli Öğrenme ile, uzmanların ödül fonksiyonu parametrelerini keşfetmekti. Bunun için, iki tane farklı Tersine Pekiştirmeli Öğrenme mekanizması oluşturduk ve sonuçlarımız geçerli ödül fonksiyonu parametreleri keşfettiğimizi önermektedir
Collective voice of experts in multilateral negotiation
No full textInspired from the ideas such as “algorithm portfolio”, “mixture of experts”, and “genetic algorithm”, this paper presents two novel negotiation strategies, which combine multiple negotiation experts to decide what to bid and what to accept during the negotiation. In the first approach namely incremental portfolio, a bid is constructed by asking each negotiation agent’s opinion in the portfolio and picking one of the suggestions stochastically considering the expertise levels of the agents. In the second approach namely crossover strategy, each expert agent makes a bid suggestion and a majority voting is used on each issue value to decide the bid content. The proposed approaches have been evaluated empirically and our experimental results showed that the crossover strategy outperformed the top five finalists of the ANAC 2016 Negotiation Competition in terms of the obtained average individual utility
Collective voice of experts in multilateral negotiation
No full textDue to copyright restrictions, the access to the full text of this article is only available via subscription.Inspired from the ideas such as “algorithm portfolio”, “mixture of experts”, and “genetic algorithm”, this paper presents two novel negotiation strategies, which combine multiple negotiation experts to decide what to bid and what to accept during the negotiation. In the first approach namely incremental portfolio, a bid is constructed by asking each negotiation agent’s opinion in the portfolio and picking one of the suggestions stochastically considering the expertise levels of the agents. In the second approach namely crossover strategy, each expert agent makes a bid suggestion and a majority voting is used on each issue value to decide the bid content. The proposed approaches have been evaluated empirically and our experimental results showed that the crossover strategy outperformed the top five finalists of the ANAC 2016 Negotiation Competition in terms of the obtained average individual utility
Emulating perceptual development in deep reinforcement learning
No full textThe process of learning in infants differs from the traditional reinforcement learning (RL) methods in several aspects. The biggest difference is that RL assumes a stationary world and an agent with fixed sensory and motor abilities. In contrast, infant development proceeds by unfolding new perceptual and motor abilities in parallel to learning. In spite of the general notion that this staged learning leads for faster and better learning in biological systems, it is not clear how such a learning mechanism can be embedded into a reinforcement learning scenario. In this study, towards this direction, we explored how an emulated perceptual development (EPD) in an RL setting can benefit the learning. As a test bed, we took the Pong game and required the RL agent to learn to play against a pre-programmed opponent by using a policy gradient based deep RL method. During learning, inspired by the progressive perceptual development in infants, the state space representation of the RL agent was changed in stages by incorporating additional information about the environment, which largely invalidated the classical RL assumption. By comparing the proposed perceptual development based learning with the performance of baseline learners, we assessed whether the benefits of developmental learning could be transferred to deep reinforcement learning systems. The results obtained suggest that a suitable perceptual development regime may improve the learning progress and yield better performing agents. © 2025 IEEE.University of Osaka ; Japan Science and Technology Agenc
Inferring effort-safety trade off in perturbed squat-to-stand task by reward parameter estimation
No full textIn this study, an inverse reinforcement learning (IRL) method is developed to estimate the parameters of a reward function that is assumed to guide the movement of a biological or artificial agent. The workings of the method is shown on the problem of estimating the effort-safety trade-off of humans during perturbed squat-to-stand motions based on their Center of Mass (COM) trajectories. The proposed method involves data generation by reinforcement learning (RL) and a novel data augmentation mechanism followed by neural network training. After the training, the neural network acts as the reward parameter estimator given the Center of Mass (COM) trajectories as input. The performance of the developed method is assessed through systematic simulation experiments, where it is shown that the parameter estimation made by our method is significantly more accurate than the baseline of an optimized template-based IRL approach. In addition, as a proof of concept, a set of human movement data is analyzed with the developed method. The results revealed that most participants acquired a strategy that ensures low effort expenditure with a safety margin, producing COM trajectories slightly away from the effort-optimal
A PARAMETRIC STUDY OF THE THERMAL PERFORMANCE OF DOUBLE SKIN FACADES AT DIFFERENT GEOGRAPHICAL LOCATIONS USING ANNUAL ENERGY SIMULATION
No full textWhile Double Skin Façades [DSF] are considered one of the advanced techniques of designing energy efficient buildings, the decision to implement a DSF is usually driven by aesthetics. The project owners and unspecialized architects generally lack the knowledge of which types of DSFs are applicable in different climate zones. In some applications, DSF designs are increasing the energy costs of buildings due to incorrect assumptions during the system selection process. This research aims to come up with result patterns to show which types of DSFs are applicable in each American Society of Heating, Refrigerating and Air-Conditioning Engineers [ASHRAE] climate zone to guide project owners and unspecialized architects during their decision making process, when they choose to design a DSF. Using Building Energy Modeling [BEM] software, various types of DSFs are analyzed at different locations. The thermal performance of the DSF is determined by comparing the building energy use data of a generic office building. The BEM tool chosen to run the analysis is Virtual Environment by Integrated Environmental Solutions [IESVE]. The different types of DSFs are created following a set of parameters such as stratification type, permissibility of airflow, and width of interstitial space. The weather data for different locations is obtained from U.S. Department of Energy [DOE] website. Although there has been much research done regarding the thermal performance of DSFs, there is a significant gap in terms of parametric and location based evaluation. Moreover, previous research tend to focus on a very specific function or type of DSF, while this study aims to create a general guide for practitioners in the decision making level of the building construction industry.M.S. in Civil Engineering, December 201
Hagadah : echah apropyadhah por lah anyadhah delah gerrah del anyo 5673
No full textAuthor can be also entered as "Yotfata, Rabbi"16 pagesPrayerbook
Patterns and Behavioural Outcomes of Antipsychotic Use among Nursing Home Residents: a Canadian and Swiss Comparison
No full textI hereby declare that I am the sole author of this thesis. This is a true copy of the thesis, including any required final revisions, as accepted by my examiners. I understand that my thesis may be made electronically available to the public. ii Background. Although antipsychotic medications are primarily intended to treat schizophrenia and psychotic symptoms in adults, they are commonly administered to nursing home residents as pharmacotherapy for “off-label ” indications such as disruptive behaviour. However, clinical trials have demonstrated limited efficacy and serious side-effects of antipsychotics among the elderly. As previous studies have reported inappropriate use in several countries, their use in nursing home residents ought to be monitored to detect and reduce inappropriate administration. Objectives. The aim of this study was a) to determine and compare prevalence rates of antipsychotic use in Ontario and Swiss nursing homes, b) to identify determinants of antipsychotics use in these two countries, by means of a cross-sectional design, and c) to investigate the impact of antipsychotic use on behaviours over time in Ontario and Swiss residents, by means of a longitudinal design
Inferring cost functions using reward parameter search and policy gradient reinforcement learning
No full textThis study focuses on inferring cost functions of obtained movement data using reward parameter search and policy gradient based Reinforcement Learning (RL). The behavior data for this task is obtained through a series of squat-to-stand movements of human participants under dynamic perturbations. The key parameter searched in the cost function is the weight of total torque used in performing the squat-to-stand action. An approximate model is used to learn squat-to-stand movements via a policy gradient method, namely Proximal Policy Optimization(PPO). A behavioral similarity metric based on Center of Mass(COM) is used to find the most likely weight parameter. The stochasticity in the training result of PPO is dealt with multiple runs, and as a result, a reasonable and a stable Inverse Reinforcement Learning(IRL) algorithm is obtained in terms of performance. The results indicate that for some participants, the reward function parameters of the experts were inferred successfully.Slovenia/ARRS -Turkey/TUBITAK bilateral collaboration ; Bogazici Resarch Fund (BAP) IMAGINE-COG++ Projec
The Factors Affecting Collaborative Building Design
No full textCollaboration is an important requisite in the multi-participant building design environment. The performance of a building as an end product is enabled via the collaborative efforts of project participants. The main objective of this paper is to identify the factors that affect the collaboration among the participants in the building design process. The related data is collected from the growing literature on collaborative design. Recent technological developments that serve a higher level of collaboration and the construction industry's efforts to improve building performance are also considered. A conceptual model is proposed to address the collaborative relationship between the participants of the building design process. The challenge is to enable collaboration by integrating the work performed by participants with diverse backgrounds, varying levels of expertise, and different perspectives. The implication of this research is that a good understanding of the factors that enhance collaboration between the parties involved in design not only enhances building performance, but also improves the competitiveness of building design firms.Peer reviewe
