You live, you learn, you forget: Continuous learning of visual places with a forgetting mechanism

Ullah, Muhammad Muneeb; Orabona, Francesco; Caputo, Barbara

doi:10.1109/iros.2009.5354588

Cited by 4 publications

(8 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When the threshold values increases, also the number of true positives start decreasing. Figure 5, right, shows the effect of using the temporal continuity ( , history factor'): for a threshold value of 0.2, we see that not only the number of true positives increases con siderably, but that up to 25 We also tested different values of �, but we did not observe any improvement in the performance, given that all the experiments we run on data collected in the same indoor environment, we believe that this result cannot be considered conclusive for an evaluation of its usefulness.…”

Section: A Experiments J: Memory-controlled Oisvmmentioning

confidence: 99%

“…The very first online algorithm to have a fixed memory "budget" and at the same time to have a relative mistake bound has been the F orgetron [7]. Within the context of semantic scene recognition, Ullah et al [25] proposed instead a random forgetting strategies, which should be more robust to possible unbalancing into the class-by class distribution of the T Ss.…”

Section: Memory Controlled Online Independent Support Vector Machinesmentioning

confidence: 99%

“…We take an discriminative approach and we build on previous work on online learning [18], [25] and confidence-based place classification [19]. Experiments on a subset of the challenging COLD database [20] show promising results.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Towards semi-supervised learning of semantic spatial concepts

Gómez¹,

Caputo

2011

2011 IEEE International Conference on Robotics and Automation

Self Cite

View full text Add to dashboard Cite

The ability of building robust semantic space representations of environments is crucial for the development of truly autonomous robots. This task, inherently connected with cognition, is traditionally achieved by training the robot with a supervised learning phase. We argue that the design of robust and autonomous systems would greatly benefit from adopting a semi-supervised online learning approach. Indeed,the support of open-ended, lifelong learning is fundamental in order to cope with the dazzling variability of the real world, and online learning provides precisely this kind of ability. Here we focus on the robot place recognition problem, and we present an online place classification algorithm that is able to detect gap in its own knowledge based on a confidence measure. For every incoming new image frame, the method is able to decide if (a) it is a known room with a familiar appearance, (b) it is a known room with a challenging appearance, or (c) it is a new, unknown room. Experiments on a subset of the challenging COLD database show the promise of our approach.

show abstract

Section: A Experiments J: Memory-controlled Oisvmmentioning

confidence: 99%

Section: Memory Controlled Online Independent Support Vector Machinesmentioning

confidence: 99%

See 1 more Smart Citation

Towards semi-supervised learning of semantic spatial concepts

Gómez¹,

Caputo

2011

2011 IEEE International Conference on Robotics and Automation

Self Cite

View full text Add to dashboard Cite

show abstract

“…In the research of place classification, although many researchers agree that the robot's space representation must at least partially overlap with human spatial concepts [17,54], there remains controversy over the selection of target class labels.…”

Section: Labelsmentioning

confidence: 99%

Towards generalization of semi-supervised place classification over generalized Voronoi graph

Shi

Kodagoda

2013

Robotics and Autonomous Systems

View full text Add to dashboard Cite

With the progress of human-robot interaction (HRI), the ability of a robot to perform highlevel tasks in complex environments is fast becoming an essential requirement. To this end, it is desirable for a robot to understand the environment at both geometric and semantic levels.Therefore in recent years, research towards place classification has been gaining in popularity. After the era of heuristic and rule-based approaches, supervised learning algorithms have been extensively used for this purpose, showing satisfactory performance levels. However, most of those approaches have only been trained and tested in the same environments and thus impede a generalized solution. In this paper, we have proposed a semi-supervised place classification over a generalized voronoi graph (SPCoGVG) which is a semi-supervised learning framework comprised of three techniques: support vector machine (SVM), conditional random field (CRF) and generalized voronoi graph (GVG), in order to improve the generalizability. The inherent problem of training CRF with partially observed data has been solved using a novel parameter estimation algorithm. The effectiveness of the proposed algorithm is validated through extensive analysis of data collected at international university environments.

show abstract

“…It can provide, for example, input for reinforcement learning during robot idle times, or a robust and persistent world model to store acquired information for a longer time [6]. A particular example we briefly present is an approach to use the database for systematic manual fault analysis.…”

Section: Introductionmentioning

confidence: 99%

A generic robot database and its application in fault analysis and performance evaluation

Niemueller

Lakemeyer

Srinivasa

2012

2012 IEEE/RSJ International Conference on Intelligent Robots and Systems

View full text Add to dashboard Cite

Abstract-During operation of robots large amounts of data are produced and processed for instance in perception, actuation, or decision making. Nowadays this data is typically volatile and disposed right after use. But this data can be valuable and useful later. Therefore we propose a database system that taps into common robot middleware to record any and all data produced at run-time. We present two examples using this data in fault analysis and performance evaluation and describe real-world experiments run on the domestic service robot HERB.

show abstract

You live, you learn, you forget: Continuous learning of visual places with a forgetting mechanism

Cited by 4 publications

References 15 publications

Towards semi-supervised learning of semantic spatial concepts

Towards semi-supervised learning of semantic spatial concepts

Towards generalization of semi-supervised place classification over generalized Voronoi graph

A generic robot database and its application in fault analysis and performance evaluation

Contact Info

Product

Resources

About