ResearchSpace
→
Browsing by Author

Browsing by Author "Rosman, Benjamin S"

Now showing items 1-20 of 40

Accelerating model learning with inter-robot knowledge transfer

Author: Makondo, Ndivhuwo ; Rosman, Benjamin S ; Hasegawa, O Date: May 2018 Online learning of a robot’s inverse dynamics model for trajectory tracking necessitates an interaction between the robot and its environment to collect training data. This is challenging for physical robots in the real world, especially for ... Read more
Action priors for learning domain invariances

Author: Rosman, Benjamin S ; Ramamoorthy, S Date: Apr 2015 An agent tasked with solving a number of different decision making problems in similar environments has an opportunity to learn over a longer timescale than each individual task. Through examining solutions to different tasks, it can uncover ... Read more
Adapting interaction environments to diverse users through online action set selection

Author: Hassan Mahmud, MM ; Rosman, Benjamin S ; Ramamoorthy, S ; Kohli, P Date: Jun 2014 Interactive interfaces are a common feature of many systems ranging from field robotics to video games. In most applications, these interfaces must be used by a heterogeneous set of users, with substantial variety in effectiveness with the ... Read more
An analysis of Monte Carlo tree search

Author: James, S ; Konidaris, G ; Rosman, Benjamin S Date: Feb 2017 Monte Carlo Tree Search (MCTS) is a family of directed search algorithms that has gained widespread attention in recent years. Despite the vast amount of research into MCTS, the effect of modifications on the algorithm, as well as the manner ... Read more
Autonomous prediction of performance-based standards for heavy vehicles

Author: Berman, R ; Benade, R ; Rosman, Benjamin S Date: Nov 2015 In most countries throughout the world, heavy vehicle use on public roads are governed by prescriptive rules, typically by imposing stringent mass and dimension limits in an attempt to control vehicle safety. A recent alternative framework ... Read more
A bayesian approach for learning and tracking switching, non-stationary opponents

Author: Hernandez-Leal, P ; Rosman, Benjamin S ; Taylor, ME ; Sucar, LE ; de Cote, EM Date: Feb 2016 In many situations, agents are required to use a set of strategies (behaviors) and switch among them during the course of an interaction. This work focuses on the problem of recognizing the strategy used by an agent within a small number of ... Read more
Bayesian policy reuse

Author: Rosman, Benjamin S ; Hawasly, M ; Ramamoorthy, S Date: Feb 2016 A long-lived autonomous agent should be able to respond online to novel instances of tasks from a familiar domain. Acting online requires `fast' responses, in terms of rapid convergence, especially when the task instance has a short duration ... Read more c b
Behavioural domain knowledge transfer for autonomous agents

Author: Rosman, Benjamin S Date: Nov 2014 An agent continuously performing different tasks in the same domain has the opportunity to learn, over the course of its operational lifetime, about the behavioural regularities afforded by the domain. This paper addresses the problem of ... Read more
Belief reward shaping in reinforcement learning

Author: Marom, O ; Rosman, Benjamin S Date: Feb 2018 A key challenge in many reinforcement learning problems is delayed rewards, which can significantly slow down learning. Although reward shaping has previously been introduced to accelerate learning by bootstrapping an agent with additional ... Read more
CHAMP: A bespoke integrated system for mobile manipulation

Author: Van Eden, Beatrice ; Rosman, Benjamin S ; Withey, Daniel J ; Ratshidaho, T ; Keaikitse, M ; Masha, D ; Kleinhans, A ; Shaik, A Date: Nov 2014 Mobile manipulation is a robotics paradigm with the potential to make major contributions to a number of important domain areas. Although some mobile manipulators are commercially available, bespoke systems can be assembled from existing and ... Read more
Context-based online policy instantiation for multiple tasks and changing environments

Author: Rosman, Benjamin S Date: 2014 This paper addresses the problem of online decision making in continually changing and complex environments, with inherent incompleteness in models of change. A fully general version of this problem is intractable but many interesting domains ... Read more
Enhancing agent safety through autonomous environment adaptation

Author: Rosman, Benjamin S ; Hayes, B ; Scassellati, B Date: Aug 2015 Exploration and self-directed learning are valuable components of early childhood development. This often comes at an unacceptable safety trade-off, as infants and toddlers are especially at risk from environmental hazards that may fundamentally ... Read more
Feature selection for domain knowledge representation through multitask learning

Author: Rosman, Benjamin S Date: Oct 2014 Representation learning is a difficult and important problem for autonomous agents. This paper presents an approach to automatic feature selection for a long-lived learning agent, which tackles the trade-off between a sparse feature set which ... Read more
Fingerprint Minutiae Extraction using Deep Learning

Author: Darlow, Luke Nicholas ; Rosman, Benjamin S Date: Oct 2017 The high variability of fingerprint data (owing to, e.g., differences in quality, moisture conditions, and scanners) makes the task of minutiae extraction challenging, particularly when approached from a stance that relies on tunable algorithmic ... Read more
G3DB: A database of successful and failed grasps with RGB-D images, point clouds, mesh models and gripper parameters

Author: Kleinhans, A ; Rosman, Benjamin S ; Michalik, M ; Tripp, B ; Detry, R Date: May 2015 Autonomous grasping is a problem that receives continuous attention from our community because it is both key to many applications, and difficult to solve. The complexity of robot grasping is counter-intuitive. For us humans, planning a grasp ... Read more
Giving advice to agents with hidden goals

Author: Rosman, Benjamin S ; Ramamoorthy, S Date: May 2014 This paper considers the problem of providing advice to an autonomous agent when neither the behavioural policy nor the goals of that agent are known to the advisor. We present an approach based on building a model of “common sense” behaviour ... Read more
Hierarchical subtask discovery with non-negative matrix factorization

Author: Earle, AC ; Saxe, AM ; Rosman, Benjamin S Date: Aug 2017 Hierarchical reinforcement learning methods offer a powerful means of planning flexible behavior in complicated domains. However, learning an appropriate hierarchical decomposition of a domain into subtasks remains a substantial challenge. ... Read more
Hierarchical subtask discovery with non-negative matrix factorization

Author: Earle, AC ; Saxe, AM ; Rosman, Benjamin S Date: Apr 2018 Hierarchical reinforcement learning methods offer a powerful means of planning flexible behavior in complicated domains. However, learning an appropriate hierarchical decomposition of a domain into subtasks remains a substantial challenge. ... Read more
Hierarchy through composition with multitask LMDPs

Author: Saxe, AM ; Earle, AC ; Rosman, Benjamin S Date: Aug 2017 Hierarchical architectures are critical to the scalability of reinforcement learning methods. Most current hierarchical frameworks execute actions serially, with macro-actions comprising sequences of primitive actions. We propose a novel ... Read more
Hyperformance: Advanced PBS Performance Prediction

Author: Berman, Robert J ; Rosman, Benjamin S ; Van Niekerk, B ; Nordengen, Paul A Date: Oct 2018 The ever increasing global freight task brings with it a number of challenges for road freight transportation. The combination of high-capacity vehicles and Performance-based Standards (PBS) is proving to be a viable and sustainable option ... Read more

Now showing items 1-20 of 40

Browse

All of ResearchSpace

Quick Links

Legislation and compliance

General Enquiries

Tel: + 27 12 841 2911
Email: callcentre@csir.co.za

Physical Address
Meiring Naudé Road
Brummeria
Pretoria
South Africa

Postal Address
PO Box 395
Pretoria 0001
South Africa

Social Connect

Resources on this site are free to download and reuse according to associated licensing provision. Please read the terms and conditions of usage of each resource.

Browsing by Author "Rosman, Benjamin S"

Browsing by Author "Rosman, Benjamin S"

Browse

All of ResearchSpace

Quick Links

Legislation and compliance

General Enquiries

Social Connect