Jonas Tjomsland's research while affiliated with University of Cambridge and other places

Publications (5)

Article
Full-text available
To date, endowing robots with an ability to assess social appropriateness of their actions has not been possible. This has been mainly due to (i) the lack of relevant and labelled data and (ii) the lack of formulations of this as a lifelong learning problem. In this paper, we address these two issues. We first introduce the Socially Appropriate Dom...
Preprint
To date, endowing robots with an ability to assess social appropriateness of their actions has not been possible. This has been mainly due to (i) the lack of relevant and labelled data, and (ii) the lack of formulations of this as a lifelong learning problem. In this paper, we address these two issues. We first introduce the Socially Appropriate Do...
Preprint
Full-text available
The intuitive collaboration of humans and intelligent robots (embodied AI) in the real-world is an essential objective for many desirable applications of robotics. Whilst there is much research regarding explicit communication, we focus on how humans and robots interact implicitly, on motor adaptation level. We present a real-world setup of a human...
Preprint
We present a robotic setup for real-world testing and evaluation of human-robot and human-human collaborative learning. Leveraging the sample-efficiency of the Soft Actor-Critic algorithm, we have implemented a robotic platform able to learn a non-trivial collaborative task with a human partner, without pre-training in simulation, and using only 30...

Citations

... Furthermore, the complexity of the datasets used for the evaluation of lifelong learning tasks is very limited (e.g., [3]) and does not reflect the richness and level of uncertainty of the stimuli that artificial agents can be exposed to in the real world. Thus, novel datasets, benchmarks, and protocols are required to evaluate long-term learning and personalization in HRI scenarios (e.g., [17,29]). In addition to designing and properly evaluating lifelong learning approaches, one must consider the philosophical, ethical, and legal implications of lifelong learning agents [2]. ...
... In 2019, Ref. [22] adopted the Soft Actor Critic reinforcement learning algorithm to build a robotic platform that the robot is capable of learning cooperative tasks with people in only 30 minutes without simulation training. Reference [23] proposed a multirobot path planning algorithm with deep q-learning combined with a convolutional neural network algorithm. ...