Filters








8,783 Hits in 2.7 sec

Curriculum Learning for Cumulative Return Maximization

Francesco Foglino, Christiano Coletto Christakou, Ricardo Luna Gutierrez, Matteo Leonetti
2019 Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence  
of cumulative return maximization.  ...  We propose a task sequencing algorithm maximizing the cumulative return, that is, the return obtained by the agent across all the learning episodes.  ...  In this paper, we propose to maximize the cumulative return instead, and propose a novel heuristic algorithm for the resulting optimization problem.  ... 
doi:10.24963/ijcai.2019/320 dblp:conf/ijcai/FoglinoCGL19 fatcat:v2za4kodzfc7tkviux4oxjdyva

An Optimization Framework for Task Sequencing in Curriculum Learning

Francesco Foglino, Christiano Coletto Christakou, Matteo Leonetti
2019 2019 Joint IEEE 9th International Conference on Development and Learning and Epigenetic Robotics (ICDL-EpiRob)  
Curriculum learning in reinforcement learning is used to shape exploration by presenting the agent with increasingly complex tasks.  ...  The idea of curriculum learning has been largely applied in both animal training and pedagogy.  ...  [29] consider the problem of task sequencing for cumulative return maximization, which is a special case of the regret metric introduced below, in which the regret is taken with respect to the optimal  ... 
doi:10.1109/devlrn.2019.8850690 dblp:conf/icdl-epirob/FoglinoCL19 fatcat:itpgca5pkvh77a7t7kupooygga

Learning to Grasp from 2.5D images: a Deep Reinforcement Learning Approach [article]

Alessia Bertugli, Paolo Galeone
2019 arXiv   pre-print
In this paper, we propose a deep reinforcement learning (DRL) solution to the grasping problem using 2.5D images as the only source of information.  ...  Unity 3D allowed us to simulate a real-world setup, where a depth camera is placed in a fixed position and the stream of images is used by our policy network to learn how to solve the task.  ...  ACKNOWLEDGMENT The authors would like to thank Dario Lodi Rizzini of the University of Parma, for several fruitful discussions and valuable advice.  ... 
arXiv:1908.03440v1 fatcat:oiwdvbiplbdctkytunvcdrm4g4

Curriculum Learning for Reinforcement Learning Domains: A Framework and Survey [article]

Sanmit Narvekar and Bei Peng and Matteo Leonetti and Jivko Sinapov and Matthew E. Taylor and Peter Stone
2020 arXiv   pre-print
Finally, we use our framework to find open problems and suggest directions for future RL curriculum learning research.  ...  In this article, we present a framework for curriculum learning (CL) in reinforcement learning, and use it to survey and classify existing CL methods in terms of their assumptions, capabilities, and goals  ...  Part of this work has taken place in the Learning Agents Research Group (LARG) at the Artificial Intelligence Laboratory, The University of Texas at Austin. LARG re-  ... 
arXiv:2003.04960v2 fatcat:iacmqeb7jjeezpo27jsnzuqb7u

Automatic Curriculum Learning For Deep RL: A Short Survey [article]

Rémy Portelas, Cédric Colas, Lilian Weng, Katja Hofmann, Pierre-Yves Oudeyer
2020 arXiv   pre-print
Automatic Curriculum Learning (ACL) has become a cornerstone of recent successes in Deep Reinforcement Learning (DRL).These methods shape the learning trajectories of agents by challenging them with tasks  ...  The ambition of this work is dual: 1) to present a compact and accessible introduction to the Automatic Curriculum Learning literature and 2) to draw a bigger picture of the current state of the art in  ...  Automatic Curriculum Learning for DRL This section formalizes the definition of ACL for Deep RL and proposes a classification.  ... 
arXiv:2003.04664v2 fatcat:lhire3htmnenfetx2ry4furgyy

Automatic Curriculum Learning through Value Disagreement [article]

Yunzhi Zhang, Pieter Abbeel, Lerrel Pinto
2020 arXiv   pre-print
Inspired by this, we propose setting up an automatic curriculum for goals that the agent needs to solve.  ...  To operationalize this idea, we introduce a goal proposal module that prioritizes goals that maximize the epistemic uncertainty of the Q-function of the policy.  ...  We also thank AWS for computational resources.  ... 
arXiv:2006.09641v1 fatcat:eozppmwgtjgs7i35jjvdmgschq

An International Survey of Gross Anatomy Courses in Chiropractic Colleges

Jennette J. Ball, Kristina L. Petrocco-Napuli, Michael P. Zumpano
2012 Journal of Chiropractic Education  
Results: Forty-four percent of the electronic surveys were returned and information was gathered for 31 institutions from public sources.  ...  Purpose: The purpose of this study is to provide the first comprehensive description of gross anatomy course design in chiropractic colleges internationally and to provide baseline data for future investigation  ...  ACKNOWLEDGMENTS The authors thank all of the survey respondents for their willingness to provide information on the anatomy curriculum at the chiropractic institutions with which they are affiliated, and  ... 
doi:10.7899/jce-12-004 pmid:23362365 pmcid:PMC3557653 fatcat:qw3mbsvp3rf7vnkx7rtzuoavka

General AI Challenge - Round One: Gradual Learning [article]

Jan Feyereisl, Matej Nikl, Martin Poliak, Martin Stransky, Michal Vlasak
2017 arXiv   pre-print
The challenge comprises of multiple rounds, with the first round focusing on gradual learning, i.e. the ability to re-use already learned knowledge for efficiently learning to solve subsequent problems  ...  We also outline a more formal description of the challenge and present a preliminary analysis of its curriculum, based on ideas from computational mechanics.  ...  A different curriculum is necessary. Curricula appropriate for a gradually learning agent also form a distribution C from which a training and an evaluation curriculum should be drawn.  ... 
arXiv:1708.05346v1 fatcat:knsyzssq3vekxnbc7ogeow3avq

Medical Student Engagement in a Virtual Learning Environment Positively Correlates with Course Performance and Satisfaction in Psychiatry

Larrilyn L. Grant, Michael J. Opperman, Brennan Schiller, Jonathan Chastain, Jennelle Durnett Richardson, Christine Eckel, Martin H. Plawecki
2021 Medical Science Educator  
The first-year medical student psychiatry curriculum was redesigned with an FC approach and subsequently altered by COVID-19 to a virtual learning environment.  ...  The FC promotes active learning and utilizes independent preparation prior to in-class sessions.  ...  Trujillo, the Assessment and Evaluation Specialist for Medical Student Education, for assistance with official end-of-course data and Dr. Dylan Powell for his help with the curriculum redesign.  ... 
doi:10.1007/s40670-021-01287-x pmid:33868773 pmcid:PMC8041389 fatcat:ikbhjai4hngvjbcwww74mwjpwq

Humanising the Design and Technology curriculum: why technology education makes us human

Matt McLain, Dawne Irving-Bell, David Wooff, David Morrison-Love
2021 figshare.com  
2019, p. 5); which may result in a resurgence of opportunities for pupils to study practical and creative subjects, such as D&T, in opposition to the perverse incentives that have led to said narrowing  ...  The aim of this paper is to reposition and reinvigorate how D&T is interpreted and enacted within the school curriculum.  ...  Both cumulative and segmented learning have their merits and problems, and the strength of the later (of which much encompasses D&T learning) is contextualised learning, which is also criticised for potentially  ... 
doi:10.6084/m9.figshare.14658420.v1 fatcat:xydvxczlb5emxmticed5vf2hyi

Meta Automatic Curriculum Learning [article]

Rémy Portelas, Clément Romac, Katja Hofmann, Pierre-Yves Oudeyer
2021 arXiv   pre-print
In such complex task spaces, it is essential to rely on some form of Automatic Curriculum Learning (ACL) to adapt the task sampling distribution to a given learning agent, instead of randomly sampling  ...  In this work, we present AGAIN, a first instantiation of Meta-ACL, and showcase its benefits for curriculum generation over classical ACL in multiple simulated environments including procedurally generated  ...  In meta-RL, agents are learning to learn to act [43] , i.e. their objective is to maximize performance on previously unseen test tasks after 0 or a few learning updates.  ... 
arXiv:2011.08463v3 fatcat:qnxqwfdvj5amvi3tjt733b5vhe

Curriculum Learning with a Progression Function [article]

Andrea Bassich, Francesco Foglino, Matteo Leonetti, Daniel Kudenko
2021 arXiv   pre-print
Curriculum Learning for Reinforcement Learning is an increasingly popular technique that involves training an agent on a sequence of intermediate tasks, called a Curriculum, to increase the agent's performance  ...  This paper introduces a novel paradigm for curriculum generation based on progression and mapping functions.  ...  The agent aims to find the optimal policy π * that maximizes the expected return.  ... 
arXiv:2008.00511v2 fatcat:hbcx7s47bjc67mcdzoamfnvwla

Evolving Rewards to Automate Reinforcement Learning [article]

Aleksandra Faust and Anthony Francis and Dar Mehta
2019 arXiv   pre-print
Many continuous control tasks have easily formulated objectives, yet using them directly as a reward in reinforcement learning (RL) leads to suboptimal policies.  ...  AutoRL, evaluated on four Mujoco continuous control tasks over two RL algorithms, shows improvements over baselines, with the the biggest uplift for more complex tasks. The video can be found at: .  ...  Acknowledgments We thank Oscar Ramirez, Rico Jonschkowski, Shane Gu, Sam Fishman, Eric Jang, Sergio Guadarrama, Sergey Levine, Brian Ichter, Hao-Tien Chiang, Jie Tan & Vincent Vanhoucke for their input  ... 
arXiv:1905.07628v1 fatcat:uycuysybq5hanljqa44nd5grz4

Curriculum Offline Imitation Learning [article]

Minghuan Liu, Hanye Zhao, Zhengyu Yang, Jian Shen, Weinan Zhang, Li Zhao, Tie-Yan Liu
2022 arXiv   pre-print
from adaptive neighboring policies with a higher return, and improves the current policy along curriculum stages.  ...  Observing that behavior cloning is able to imitate neighboring policies with less data, we propose Curriculum Offline Imitation Learning (COIL), which utilizes an experience picking strategy for imitating  ...  We sincerely thank the reviewers for helpful feedback.  ... 
arXiv:2111.02056v2 fatcat:d7obnsol55bdlnec7gim5juzuq

The Canary in the Mine: Remote Learning in the Time of COVID-19

David Coker
2020 Journal of Curriculum and Teaching  
National organizations in the United States issued policy proposals for returning to school during the COVID-19 pandemic.  ...  A case study using an individual school district examined the impact on learning within the framework of the policies. Recommendations to improve online and remote learning follow.  ...  How will schools know students are learning? What will be different from Spring 2020? Results could be useful in developing policies and procedures to maximize student learning.  ... 
doi:10.5430/jct.v9n3p76 fatcat:c7j35g5jqrgbheg46yoc6sygf4
« Previous Showing results 1 — 15 out of 8,783 results