Fig 5 - uploaded by Jure Zabkar
Content may be subject to copyright.
Source publication
We present a novel approach to embodied learning of qualitative models. We introduce algorithm STRUDEL that enables an autonomous robot to discover new concepts by performing experiments in its environment. The robot collects data about its actions and its observations of the environment. From the obtained data, the robot learns qualitative descrip...
Context in source publication
Context 1
... hands, squeezing them tightly so they did not drop out of its grip. The robot perceives the shape of the object using its overhead vision system, like the object's location. The shape is considered to be an observation, not background knowledge, and the robot does not know if it can change or not. This time the robot was given an action named put (Fig. 5) to experiment with. This action is performed using two selected objects. The robot picks up the first object with both hands and places it on top of the other selected object. The robot has to carry the picked up object in its hands to the second object to achieve this. While walking around and carrying objects, it again avoids bumping ...
Similar publications
It has been an ongoing societal quest to collect more and more data; generation of knowledge from it, however, has been very disproportionate. Even massive amount of collected data is no indication that their understanding is a simple process of calculating probabilities; selection of what to collect must be proactive and done by the 'best minds' b...
Citations
... Leban et al. [23] used such an approach with a robot that learned the concept of movability and a notion of obstacle by experimentally interacting with objects. Košmerlj et al. [18] proposed certain improvements to reduce the computational complexity of learning. Mugan and Kuipers [27] developed an unsupervised learning algorithm called QLAP (Qualitative Learner of Action and Perception) that allows learning of action policies through autonomous exploration by motor bubbling or by explicit coaching.Šuc et al. [40,41] proposed learning qualitative trees to obtain models that intuitively explain how the system works. ...
Qualitative modeling allows autonomous agents to learn comprehensible control models, formulated in a way that is close to human intuition. By abstracting away certain numerical information, qualitative models can provide better insights into operating principles of a dynamic system in comparison to traditional numerical models. We show that qualitative models, learned from numerical traces, contain enough information to allow motion planning and path following. We demonstrate our methods on the task of flying a quadcopter. A qualitative control model is learned through motor babbling. Training is significantly faster than training times reported in papers using reinforcement learning with similar quadcopter experiments. A qualitative collision-free trajectory is computed by means of qualitative simulation, and executed reactively while dynamically adapting to numerical characteristics of the system. Experiments have been conducted and assessed in the V-REP robotic simulator.
... Some of the most relevant works will be described as follows. Multiple concept learning with predicate invention in robotics domains has been addressed in Hyper ( Kosmerlj, Bratko, & Zabkar, 2011;Leban, Zabkar, & Bratko, 2008 ). Hyper starts with background knowledge and can perform predicate invention adding placeholder predicates (predicates which are being invented) to learn the target predicates. ...
A truly autonomous artificial intelligence agent should be able to drive its own learning process. That is, decide what to explore and what to learn, identifying what constitutes potential useful data as examples of concepts or what strategy to follow to solve a new task. Different efforts have been developed in machine learning towards this aim. Approaches that introduce new concepts, like predicate invention in Inductive Logic Programming (ILP) techniques, normally require the selection of examples by the user. Techniques that learn behavior policies through exploration like Reinforcement Learning (RL) with intrinsic motivation, to guide the agent into interesting areas to discover new goals, assume that all the states and actions are predefined in advance. In this paper, we describe a system, called ADC, that combines techniques from ILP with predicate invention and RL with intrinsic motivation to discover new concepts, states and actions to learn behavior policies. ADC drives its own learning process, collecting its own examples for autonomously learning concepts. These new concepts can be used to describe its environment and define new states and actions used to learn behaviors to solve tasks. We show the effectiveness of our approach in simulated robotics environments.
... In ML, Inductive Logic programming (ILP) (Muggleton, 1991) has been successful in the learning of concepts because of its powerful representation based on logic and its robust inductive methods. ILP can also provide mechanisms to add new vocabulary to the initial BK as demand-driven predicate invention (Muggleton & Road, 1994;Stahl, 1996;Wrobel, 1994;Li et al., 2008;Stanley & Domingos, 2007;Kosmerlj et al., 2011). Reinforcement learning (RL) (Sutton & Barto, 1998) has been useful to acquire skills from experience when external rewards are provided. ...
How to represent adequate knowledge for solving a problem is a challenging task in the design of artificial intelligent agents. Ideally we would like an agent to create its own representations. In this paper, we describe the ADC (Automatic Discovery of Concepts) algorithm for learning re-lational concepts and actions to represent states of the environment and solve tasks, using induc-tive logic programming and reinforcement learning. Our experiments show that an intrinsic motivation function and automatically learning new representations can reduce the learning time and produce useful concepts for solving a task.
Automatic discovery of concepts has been an elusive area in machine learning. In this paper, we describe a system, called ADC, that automatically discovers concepts in a robotics domain, performing predicate invention. Unlike traditional approaches of concept discovery, our approach automatically finds and collects instances of potential relational concepts. An agent, using ADC, creates an incremental graph-based representation with the information it gathers while exploring its environment, from which common sub-graphs are identified. The subgraphs discovered are instances of potential relational concepts which are induced with Inductive Logic Programming and predicate invention. Several concepts can be induced concurrently and the learned concepts can form arbitrarily hierarchies. The approach was tested for learning concepts of polygons, furniture, and floors of buildings with a simulated robot and compared with concepts suggested by users.
A reliable human skin detection method that is adaptable to different human skin colors and illumination conditions is essential for better human skin segmentation. Even though different human skin-color detection solutions have been successfully applied, they are prone to false skin detection and are not able to cope with the variety of human skin colors across different ethnic. Moreover, existing methods require high computational cost. In this paper, we propose a novel human skin detection approach that combines a smoothed 2-D histogram and Gaussian model, for automatic human skin detection in color image(s). In our approach, an eye detector is used to refine the skin model for a specific person. The proposed approach reduces computational costs as no training is required, and it improves the accuracy of skin detection despite wide variation in ethnicity and illumination. To the best of our knowledge, this is the first method to employ fusion strategy for this purpose. Qualitative and quantitative results on three standard public datasets and a comparison with state-of-the-art methods have shown the effectiveness and robustness of the proposed approach.
