Explainable Post-Occupancy Evaluation Using a Humanoid Robot

Reinforcement Learning (RL) methods provide a solution for decision-making problems under uncertainty. An agent finds a suitable policy through a reward function by interacting with a dynamic environment. However, for complex and large problems it is very difficult to specify and tune the reward function. Inverse Reinforcement Learning (IRL) may mitigate this problem by learning the reward function through expert demonstrations. This work exploits an IRL method named Max-Margin Algorithm (MMA) to learn the reward function for a robotic navigation problem. The learned reward function reveals the demonstrated policy (expert policy) better than all other policies. Results show that this method has better convergence and learned reward functions through the adopted method represents expert behavior more efficiently.

Section: Related Workmentioning

confidence: 99%

“…= w . μ π (16) Algorithm-1 is used to find a weight vector w that minimizes the difference between the expert feature expectations μ E and the estimated feature expectations μπ . Let consider an MDP without reward MDP/R.…”

Section: R(s A)mentioning

confidence: 99%

Inverse Reinforcement Learning Through Max-Margin Algorithm

Shah

Coronato

2021

“…ML is an application of AI that focuses on learning and improving itself from experience and without being explicitly programmed. ML emphasizes on developing algorithms that can access data and use it for self-learning [1,2,3] in an intelligent environments [4,5,6]. We are dealing with a certain number of sensors, which enable the IE [7] to be aware of the user's current action and goal.…”

Section: Introductionmentioning

confidence: 99%

An Overview of Inverse Reinforcement Learning Techniques

Shah

Pietro

2021

In decision-making problems reward function plays an important role in finding the best policy. Reinforcement Learning (RL) provides a solution for decision-making problems under uncertainty in an Intelligent Environment (IE). However, it is difficult to specify the reward function for RL agents in large and complex problems. To counter these problems an extension of RL problem named Inverse Reinforcement Learning (IRL) is introduced, where reward function is learned from expert demonstrations. IRL is appealing for its potential use to build autonomous agents, capable of modeling others, deprived of compromising in performance of the task. This approach of learning by demonstrations relies on the framework of Markov Decision Process (MDP). This article elaborates original IRL algorithms along with their close variants to mitigate challenges. The purpose of this paper is to highlight an overview and theoretical background of IRL in the field of Machine Learning (ML) and Artificial Intelligence (AI). We presented a brief comparison between different variants of IRL in this article.

“…reinforcement learning [23], neural network [21] the goal of Artificial Intelligence (AI) has become a step closer. AI has important application in diverse fields including:healthcare [12], [22], robotics and autonomous control, vision enhancing method for low vision impairments [19], natural language processing, dynamic normative environments [31], risk management [26], intelligent environments [10], games and self-organized system [11], ambient assisted living techniques to improve the quality of life of elderly [13], Social humanoid robot [9] can help to monitor indoor environmental quality [29] and distributed fuzzy system able to infer in real-time critical situations [30]. In this paper, we propose a novel approach for the diagnostic prediction of Breast Cancer by careful feature selection and data handling.…”

Section: Introductionmentioning

confidence: 99%

Prediction of Breast Cancer Using AI-Based Methods

Aamir¹,

Rahim²,

Bashir

et al. 2021

Breast cancer has made its mark as the primary cause of female deaths and disability worldwide, making it a significant health problem. However, early diagnosis of breast cancer can lead to its effective treatment. The relevant diagnostic features available in the patient’s medical data may be used in an effective way to diagnose, categorize and classify breast cancer. Considering the importance of early detection of breast cancer in its effective treatment, it is important to accurately diagnose and classify breast cancer using diagnostic features present in available data. Automated techniques based on machine learning are an effective way to classify data for diagnosis. Various machine learning based automated techniques have been proposed by researches for early prediction/diagnosis of breast cancer. However, due to the inherent criticalities and the risks coupled with wrong diagnosis, there is a dire need that the accuracy of the predicted diagnosis must be improved. In this paper, we have introduced a novel supervised machine learning based approach that embodies Random Forest, Gradient Boosting, Support Vector Machine, Artificial Neural Network and Multilayer Perception methods. Experimental results show that the proposed framework has achieved an accuracy of 99.12%. Results obtained after the process of feature selection indicate that both preprocessing methods and feature selection increase the success of the classification system.