2019 Joint IEEE 9th International Conference on Development and Learning and Epigenetic Robotics (ICDL-EpiRob)最新文献

Training-ValueNet: Data Driven Label Noise Cleaning on Weakly-Supervised Web Images Training-ValueNet:弱监督Web图像上数据驱动的标签噪声清除

2019 Joint IEEE 9th International Conference on Development and Learning and Epigenetic Robotics (ICDL-EpiRob) Pub Date : 2019-09-30 DOI: 10.1109/DEVLRN.2019.8850689

Luka Smyth, D. Kangin, N. Pugeault

{"title":"Training-ValueNet: Data Driven Label Noise Cleaning on Weakly-Supervised Web Images","authors":"Luka Smyth, D. Kangin, N. Pugeault","doi":"10.1109/DEVLRN.2019.8850689","DOIUrl":"https://doi.org/10.1109/DEVLRN.2019.8850689","url":null,"abstract":"Manually labelling new datasets for image classification remains expensive and time-consuming. A promising alternative is to utilize the abundance of images on the web for which search queries or surrounding text offers a natural source of weak supervision. Unfortunately the label noise in these datasets has limited their use in practice. Several methods have been proposed for performing unsupervised label noise cleaning, the majority of which use outlier detection to identify and remove mislabeled images. In this paper, we argue that outlier detection is an inherently unsuitable approach for this task due to major flaws in the assumptions it makes about the distribution of mislabeled images. We propose an alternative approach which makes no such assumptions. Rather than looking for outliers, we observe that mislabeled images can be identified by the detrimental impact they have on the performance of an image classifier. We introduce training-value as an objective measure of the contribution each training example makes to the validation loss. We then present the training-value approximation network (Training-ValueNet) which learns a mapping between each image and its training-value. We demonstrate that by simply discarding images with a negative training-value, Training-ValueNet is able to significantly improve classification performance on a held-out test set, outperforming the state of the art in outlier detection by a large margin.","PeriodicalId":318973,"journal":{"name":"2019 Joint IEEE 9th International Conference on Development and Learning and Epigenetic Robotics (ICDL-EpiRob)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128518929","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 4

Learning to Parse Grounded Language using Reservoir Computing 学习使用库计算解析基础语言

2019 Joint IEEE 9th International Conference on Development and Learning and Epigenetic Robotics (ICDL-EpiRob) Pub Date : 2019-08-19 DOI: 10.1109/DEVLRN.2019.8850718

Xavier Hinaut, Michael Spranger

{"title":"Learning to Parse Grounded Language using Reservoir Computing","authors":"Xavier Hinaut, Michael Spranger","doi":"10.1109/DEVLRN.2019.8850718","DOIUrl":"https://doi.org/10.1109/DEVLRN.2019.8850718","url":null,"abstract":"Recently new models for language processing and learning using Reservoir Computing have been popular. However, these models are typically not grounded in sensorimotor systems and robots. In this paper, we develop a model of Reservoir Computing called Reservoir Parser (ResPars) for learning to parse Natural Language from grounded data coming from humanoid robots. Previous work showed that ResPars is able to do syntactic generalization over different sentences (surface structure) with the same meaning (deep structure). We argue that such ability is key to guide linguistic generalization in a grounded architecture. We show that ResPars is able to generalize on grounded compositional semantics by combining it with Incremental Recruitment Language (IRL). Additionally, we show that ResPars is able to learn to generalize on the same sentences, but not processed word by word, but as an unsegmented sequence of phonemes. This ability enables the architecture to not rely only on the words recognized by a speech recognizer, but to process the sub-word level directly. We additionally test the model's robustness to word error recognition.","PeriodicalId":318973,"journal":{"name":"2019 Joint IEEE 9th International Conference on Development and Learning and Epigenetic Robotics (ICDL-EpiRob)","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-08-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126574807","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 6

Utilizing Pragmatic Frames as an analytical tool for children's performance during word learning 运用语用框架分析儿童词汇学习表现

2019 Joint IEEE 9th International Conference on Development and Learning and Epigenetic Robotics (ICDL-EpiRob) Pub Date : 2019-08-01 DOI: 10.1109/DEVLRN.2019.8850714

K. Rohlfing, Angela Grimminger

引用次数: 3

Action Selection Based on Prediction for Robot Planning 基于预测的机器人规划动作选择

2019 Joint IEEE 9th International Conference on Development and Learning and Epigenetic Robotics (ICDL-EpiRob) Pub Date : 2019-08-01 DOI: 10.1109/DEVLRN.2019.8850676

Mengxi Nie, D. Luo, Tianlin Liu, Xihong Wu

引用次数: 1

Hindsight Experience Replay With Experience Ranking 带有经验排名的后见之明经验回放

2019 Joint IEEE 9th International Conference on Development and Learning and Epigenetic Robotics (ICDL-EpiRob) Pub Date : 2019-08-01 DOI: 10.1109/DEVLRN.2019.8850705

Hai V. Nguyen, H. La, M. Deans

{"title":"Hindsight Experience Replay With Experience Ranking","authors":"Hai V. Nguyen, H. La, M. Deans","doi":"10.1109/DEVLRN.2019.8850705","DOIUrl":"https://doi.org/10.1109/DEVLRN.2019.8850705","url":null,"abstract":"Reinforcement Learning (RL) algorithms face difficulties when dealing with robotic tasks in sparse reward settings and as a result, they often require millions of interactions with the environment to learn successfully. A recent algorithm Hindsight Experience Replay (HER) was introduced to tackle this difficulty by adding virtual goals and therefore increase significantly the sample-efficiency by learning in transitions when the robot does not achieve the original goal. However, these additional goals are sampled randomly from each episode batch of transitions, which might have no relationship with the original goal. This might make learning with the original goal slower due to the bad influence of irrelevant virtual goals. In this paper, we address this issue by applying experience ranking (ER) to these additional goals. We first compare each sampled virtual goal and the original goal and then compare the difference with a threshold. Transitions in which the robot achieves a virtual goal that is not close to the original goal are filtered out, and the remaining are used for training the policy. The improvement in learning performance is validated in four simulated robotic tasks. The experiment results show significant improvement in terms of the learning speed and robustness.","PeriodicalId":318973,"journal":{"name":"2019 Joint IEEE 9th International Conference on Development and Learning and Epigenetic Robotics (ICDL-EpiRob)","volume":"127 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123227487","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 17

Modeling Self-organized Emergence of Perspective In/variant Mirror Neurons in a Robotic System 机器人系统中透视/变异镜像神经元的自组织涌现建模

2019 Joint IEEE 9th International Conference on Development and Learning and Epigenetic Robotics (ICDL-EpiRob) Pub Date : 2019-08-01 DOI: 10.1109/DEVLRN.2019.8850692

Jakub Pospíchal, I. Farkaš, Matej Pechác, Kristína Malinovská

{"title":"Modeling Self-organized Emergence of Perspective In/variant Mirror Neurons in a Robotic System","authors":"Jakub Pospíchal, I. Farkaš, Matej Pechác, Kristína Malinovská","doi":"10.1109/DEVLRN.2019.8850692","DOIUrl":"https://doi.org/10.1109/DEVLRN.2019.8850692","url":null,"abstract":"A major role attributed to mirror neurons, according to the direct matching hypothesis, is to mediate the link between an observed action and agent's own motor repertoire, to provide understanding “from inside”. The mirror neurons gave rise to various models but one of the issues not tackled by them is the perspective in/variance. Neurons in STS visual areas can be either perspective selective or invariant and the same variability was later also discovered in premotor F5 area in macaques, showing the existence of different types of mirror neurons regarding their perspective selectivity. We model this as an emergent phenomenom using the data from the simulated iCub robot, that learns to reach for objects with three types of grasp. The neural network model learns in two phases. First, the motor (F5) and visual (STS) modules are trained in parallel to self-organize modal maps using the corresponding data sequences from the self-perspective. Then, F5 area is retrained using the output from the pretrained STS module, to acquire the mirroring property. Using the optimized model hyperparameters found by grid search, we show that our model fits very well empirical observations, by showing how neurons with various degrees of perspective selectivity emerge in the F5 map.","PeriodicalId":318973,"journal":{"name":"2019 Joint IEEE 9th International Conference on Development and Learning and Epigenetic Robotics (ICDL-EpiRob)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123551625","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 1

Recognition in Human-Robot Interaction: The Gateway to Engagement 人机交互中的识别:参与的门户

2019 Joint IEEE 9th International Conference on Development and Learning and Epigenetic Robotics (ICDL-EpiRob) Pub Date : 2019-08-01 DOI: 10.1109/DEVLRN.2019.8850691

Ingar Brinck, C. Balkenius

引用次数: 4

Online Associative Multi-Stage Goal Babbling Toward Versatile Learning of Sensorimotor Skills 面向感觉运动技能多用途学习的在线联想多阶段目标学习

2019 Joint IEEE 9th International Conference on Development and Learning and Epigenetic Robotics (ICDL-EpiRob) Pub Date : 2019-08-01 DOI: 10.1109/DEVLRN.2019.8850707

Rania Rayyes, Jochen J. Steil

{"title":"Online Associative Multi-Stage Goal Babbling Toward Versatile Learning of Sensorimotor Skills","authors":"Rania Rayyes, Jochen J. Steil","doi":"10.1109/DEVLRN.2019.8850707","DOIUrl":"https://doi.org/10.1109/DEVLRN.2019.8850707","url":null,"abstract":"We develop an online learning scheme inspired by the versatility of the human learning system to bootstrap several sensorimotor skills in “Learning while Behaving” fashion. Our proposed scheme is able to represent multiple coordination styles to handle assigned tasks flexibly. We have four main contributions in this paper. First, we propose a novel online learning scheme to learn several robot models simultaneously, online, from scratch and in a plain exploratory fashion. Second, we develop an incremental online associative radial basis function network which is constructed from scratch to solve the learned mapping ambiguities, e.g., redundancy, dynamically based on the current robot state. Third, we combine both proposed schemes to inherit their advantages in Associative Multi-Stage Goal Babbling. Fourth, we propose a parameter-sharing technique to increase efficiency and speed up the online learning process. All the proposed methods are evaluated in different illustrative experiments. They demonstrate promising performance with sufficient accuracy and a reasonable number of samples.","PeriodicalId":318973,"journal":{"name":"2019 Joint IEEE 9th International Conference on Development and Learning and Epigenetic Robotics (ICDL-EpiRob)","volume":"116 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116048926","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 4

Toddlers' Hands Organize Parent-Toddler Attention across Different Social Contexts 幼儿的手在不同的社会环境中组织父母-幼儿的注意力

2019 Joint IEEE 9th International Conference on Development and Learning and Epigenetic Robotics (ICDL-EpiRob) Pub Date : 2019-08-01 DOI: 10.1109/DEVLRN.2019.8850682

Steven L. Elmlinger, Sumarga H. Suanda, Linda B. Smith, Chen Yu

引用次数: 3

Behavioral Repertoire via Generative Adversarial Policy Networks 基于生成对抗策略网络的行为库

2019 Joint IEEE 9th International Conference on Development and Learning and Epigenetic Robotics (ICDL-EpiRob) Pub Date : 2019-08-01 DOI: 10.1109/DEVLRN.2019.8850727

Marija Jegorova, S. Doncieux, Timothy M. Hospedales

{"title":"Behavioral Repertoire via Generative Adversarial Policy Networks","authors":"Marija Jegorova, S. Doncieux, Timothy M. Hospedales","doi":"10.1109/DEVLRN.2019.8850727","DOIUrl":"https://doi.org/10.1109/DEVLRN.2019.8850727","url":null,"abstract":"Learning algorithms are enabling robots to solve increasingly challenging real-world tasks. These approaches often rely on demonstrations and reproduce the behavior shown. Unexpected changes in the environment may require using different behaviors to achieve the same effect, for instance to reach and grasp an object in changing clutter. An emerging paradigm addressing this robustness issue is to learn a diverse set of successful behaviors for a given task, from which a robot can select the most suitable policy when faced with a new environment. In this paper, we explore a novel realization of this vision by learning a generative model over policies. Rather than learning a single policy, or a small fixed repertoire, our generative model for policies compactly encodes an unbounded number of policies and allows novel controller variants to be sampled. Leveraging our generative policy network, a robot can sample novel behaviors until it finds one that works for a new environment. We demonstrate this idea with an application of robust ball-throwing in the presence of obstacles. We show that this approach achieves a greater diversity of behaviors than an existing evolutionary approach, while maintaining good efficacy of sampled behaviors, allowing a Baxter robot to hit targets more often when ball throwing in the presence of obstacles.","PeriodicalId":318973,"journal":{"name":"2019 Joint IEEE 9th International Conference on Development and Learning and Epigenetic Robotics (ICDL-EpiRob)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132361240","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 11