scholarly journals Continuous learning of emergent behavior in robotic matter

2021 ◽  
Vol 118 (21) ◽  
pp. e2017015118
Author(s):  
Giorgio Oliveri ◽  
Lucas C. van Laake ◽  
Cesare Carissimo ◽  
Clara Miette ◽  
Johannes T. B. Overvelde
Keyword(s):  
Learning Strategy ◽  
Dynamic Environment ◽  
Emergent Behavior ◽  
Learning Approach ◽  
Robotic Platform ◽  
Optimal Behavior ◽  
New Class ◽  
Autonomous Behavior ◽  
Direct Modeling ◽  
Scalable Learning

One of the main challenges in robotics is the development of systems that can adapt to their environment and achieve autonomous behavior. Current approaches typically aim to achieve this by increasing the complexity of the centralized controller by, e.g., direct modeling of their behavior, or implementing machine learning. In contrast, we simplify the controller using a decentralized and modular approach, with the aim of finding specific requirements needed for a robust and scalable learning strategy in robots. To achieve this, we conducted experiments and simulations on a specific robotic platform assembled from identical autonomous units that continuously sense their environment and react to it. By letting each unit adapt its behavior independently using a basic Monte Carlo scheme, the assembled system is able to learn and maintain optimal behavior in a dynamic environment as long as its memory is representative of the current environment, even when incurring damage. We show that the physical connection between the units is enough to achieve learning, and no additional communication or centralized information is required. As a result, such a distributed learning approach can be easily scaled to larger assemblies, blurring the boundaries between materials and robots, paving the way for a new class of modular “robotic matter” that can autonomously learn to thrive in dynamic or unfamiliar situations, for example, encountered by soft robots or self-assembled (micro)robots in various environments spanning from the medical realm to space explorations.

scholarly journals A deep learning approach to capture the essence of Candida albicans morphologies

2021 ◽  
Author(s):  
Van Bettauer ◽  
Anna CBP Costa ◽  
Raha Parvizi Omran ◽  
Samira Massahi ◽  
Eftyhios Kirbizakis ◽  
...  
Keyword(s):  
Deep Learning ◽  
Learning Strategy ◽  
Developmental Trajectories ◽  
Generative Adversarial Networks ◽  
Learning Approach ◽  
Adversarial Networks ◽  
Latent Space ◽  
Contrast Microscopy ◽  
Opportunistic Human Pathogen ◽  
Community Meeting

We present deep learning-based approaches for exploring the complex array of morphologies exhibited by the opportunistic human pathogen C. albicans. Our system entitled Candescence automatically detects C. albicans cells from Differential Image Contrast microscopy, and labels each detected cell with one of nine vegetative, mating-competent or filamentous morphologies. The software is based upon a fully convolutional one-stage object detector and exploits a novel cumulative curriculum-based learning strategy that stratifies our images by difficulty from simple vegetative forms to more complex filamentous architectures. Candescence achieves very good performance on this difficult learning set which has substantial intermixing between the predicted classes. To capture the essence of each C. albicans morphology, we develop models using generative adversarial networks and identify subcomponents of the latent space which control technical variables, developmental trajectories or morphological switches. We envision Candescence as a community meeting point for quantitative explorations of C. albicans morphology.

Synthetic Worlds, Synthetic Strategies

2011 ◽  
pp. 182-197
Author(s):  
Elif Ayiter
Keyword(s):  
Transformative Learning ◽  
Virtual Environment ◽  
Visual Arts ◽  
Teaching Strategies ◽  
Learning Strategy ◽  
Role Play ◽  
Three Dimensional ◽  
Learning Approach ◽  
Visual Identity ◽  
Definition Of

This text will attempt to delineate the underlying theoretical premises and the definition of the output of an immersive learning approach pertaining to the visual arts to be implemented in online, three dimensional synthetic worlds. Deviating from the prevalent practice of the replication of physical art studio teaching strategies within a virtual environment, the author proposes instead to apply the fundamental tenets of Roy Ascott’s “Groundcourse”, in combination with recent educational approaches such as “Transformative Learning” and “Constructionism”. In an amalgamation of these educational approaches with findings drawn from the fields of Metanomics, Ludology, Cyberpsychology and Presence Studies, as well as an examination of creative practices manifest in the metaverse today, the formulation of a learning strategy for creative enablement unique to online, three dimensional synthetic worlds; one which will focus upon “Play” as well as Role Play, virtual Assemblage and the visual identity of the avatar within the pursuits, is being proposed in this chapter.

scholarly journals A Critical Review on Blended Learning Versus Traditional Lecture Method

2021 ◽  
Vol 13 (2) ◽  
pp. 62-76
Author(s):  
Muhammad Hafeez
Keyword(s):  
Blended Learning ◽  
Critical Review ◽  
Learning Strategy ◽  
Research Articles ◽  
Learning Approach ◽  
Review Of Literature ◽  
Traditional Lecture ◽  
Effective Learning ◽  
Lecture Method ◽  
Information And Communication

From the beginning of 21st century, the leaning stratigies have been changed from traditional to information and communication based. A critical review of published articles about blended and traditional leaning stratigies has been conducted to highlight the importance and significance of both learning stratigies. Thirty-six (36) research articles published in various databases in various disciplines have been selected for review.  The review of literature showed that in most of the studies, the blended learning strategy proved to be more effective learning strategy against the traditional lecture method. From thirty-six published articles reviewed, twenty-five studies showed a statistically more significance value in blended learning approach for academic achievement, critical and creative skills in various disciplines. So, on the basis of this study, it is strongly recommended that blended learning strategy must be applied to achieve high academic and professional results.    

scholarly journals Teaching Critical Reflection to Graduate Students

2014 ◽  
Vol 7 (1) ◽  
pp. 56 ◽  
Author(s):  
Gavan Peter Longley Watson ◽  
Natasha Kenny
Keyword(s):  
Graduate Students ◽  
Critical Reflection ◽  
Learning Strategy ◽  
University Teaching ◽  
Learning Approach ◽  
Graduate Course ◽  
Applied Learning ◽  
Post Secondary ◽  
Life Long Learning ◽  
The University

Critical reflection is a highly valued and widely applied learning approach in higher education. There are many benefits associated with engaging in critical reflection, and it is often integrated into the design of graduate-level courses on university teaching, as a life-long learning strategy to help ensure that learners build their capacity as critical reflective teaching practitioners. Despite its broad application and learning benefits, students often find the process of engaging in critical reflection inherently challenging. This paper explores the challenge associated with incorporating critical reflection into a graduate course on University Teaching at the University of Guelph. Strategies for effectively incorporating critical reflection are presented, based largely on Arsonson’s (2011) framework for teaching critical reflection and the outcomes of a workshop offered at the 2013 STLHE Conference. The strategies discussed have multi-disciplinary relevance, and can be broadly applied to improve how critical reflection is incorporated into post-secondary courses.

scholarly journals Dealing with Drift Uncertainty: A Bayesian Learning Approach

Risks ◽  
2019 ◽  
Vol 7 (1) ◽  
pp. 5 ◽  
Author(s):  
Carmine De Franco ◽  
Johann Nicolle ◽  
Huyên Pham
Keyword(s):  
Bayesian Learning ◽  
Learning Strategy ◽  
Value Added ◽  
Learning Approach ◽  
Practical Procedure ◽  
Asset Classes ◽  
Two Parameters ◽  
Markowitz Problem ◽  
Drift Uncertainty ◽  
Smart Beta

One of the main challenges investors have to face is model uncertainty. Typically, the dynamic of the assets is modeled using two parameters: the drift vector and the covariance matrix, which are both uncertain. Since the variance/covariance parameter is assumed to be estimated with a certain level of confidence, we focus on drift uncertainty in this paper. Building on filtering techniques and learning methods, we use a Bayesian learning approach to solve the Markowitz problem and provide a simple and practical procedure to implement optimal strategy. To illustrate the value added of using the optimal Bayesian learning strategy, we compare it with an optimal nonlearning strategy that keeps the drift constant at all times. In order to emphasize the prevalence of the Bayesian learning strategy above the nonlearning one in different situations, we experiment three different investment universes: indices of various asset classes, currencies and smart beta strategies.

Multi-Robot Cooperative Transportation of Objects Using Modified Q-Learning

2010 ◽  
Author(s):  
Pallege Gamini Dilupa Siriwardana ◽  
Clarence de Silva
Keyword(s):  
Autonomous Robots ◽  
Dynamic Environment ◽  
Learning Approach ◽  
Robot System ◽  
Q Learning ◽  
Time Performance ◽  
Object Pose Estimation ◽  
Object Transportation ◽  
Blob Tracking ◽  
Multi Robot

In cooperative multi-robot object transportation, several autonomous robots navigate cooperatively in either a static or a dynamic environment to transport an object to a goal location and orientation. The environment may consist of both fixed and removable obstacles and it will be subject to uncertainty and unforeseen changes within the environment. More than one robot may be required for handling heavy and large objects. This paper presents a modified Q-learning approach for object transportation utilizing cooperative and autonomous multiple mobile robots. A modified version of Q-learning is presented, which employs for effective robot coordination. A solution to the action selection conflicts of the robots is presented, which helps to improve the real time performance and robustness of the system. As required in the task, the paper presents an algorithm for object pose estimation, by utilizing the laser range finder and color blob tracking. The developed techniques are implemented in a multi-robot system in laboratory. Experimental results are presented to demonstrate the effectiveness of the developed multi-robot system and its underlying methodologies.

The CPS Strategy

2019 ◽  
pp. 110-137
Author(s):  
Diana Carolina Durán-Bautista
Keyword(s):  
Foreign Language ◽  
Educational Research ◽  
Quantitative Data ◽  
Learning Strategy ◽  
Learning Approach ◽  
Flipped Learning ◽  
Online Surveys ◽  
Qualitative And Quantitative ◽  
Teachers Perspectives

This educational research-based chapter describes, analyzes, and evaluates the implementation of the class preparation session (The CPS) as a flipped learning strategy in an undergraduate English as a foreign language (EFL) program in Colombia. About 3000 students are enrolled in the EFL program each semester and an average of 40 teachers are in charge of teaching the courses. The chapter describes the process of implementing the CPS strategy and evaluates its efficacy from students' and teachers' perspectives. Participants were requested to take online surveys with the purpose of collecting qualitative and quantitative data about the strengths of the CPS and the challenges encountered its resources. The processes described in this chapter could be adjusted to be used in other institutions. The issues discussed might provide several teaching prospects for teachers and for programs' stakeholders interested in implementing the flipped learning approach.

Autonomous Formation of Transportation Order under Dynamical Environment

2000 ◽  
Vol 12 (4) ◽  
pp. 494-500 ◽  
Author(s):  
Toshimitsu Higashi ◽  
◽  
Kosuke Sekiyama ◽  
Toshio Fukuda ◽  
◽  
...  
Keyword(s):  
Autonomous System ◽  
Large Scale ◽  
Dynamic Environment ◽  
Transportation System ◽  
Global Evaluation ◽  
Evaluation Function ◽  
Environment Changes ◽  
Dynamically Reconfigurable ◽  
Relative Evaluation ◽  
Autonomous Behavior

This paper proposes a system, that realizes collective autonomous behavior such as an autonomous conveyance order formation in the AGV (Auto Guided Vehicle) transportation system. We attempt to deal with a large-scale distributed autonomous system in a dynamic environment feasibly. However, if we use a global evaluation function in order to control each agent, it is necessary to rewrite the global evaluation function of the system whenever the environment changes. If we use such a method, the system cannot be called a real distributed autonomous system. In this paper, we propose two ideas in order to realize dynamically reconfigurable formation in the dynamic environment, namely, learning based on the agent's own action and interaction with other agents by relative evaluation. By use of these ideas, it is shown that dynamically reconfigurable formation emerges as an autonomous conveyance order formation of AGV transportation in the dynamic environment.

Sign in / Sign up

Export Citation Format

Share Document