Biography
Dr Oiwi Parker Jones is a Postdoctoral Researcher in the Applied Artificial Intelligence Lab (A2I) and a Hugh Price Research Fellow at Jesus College. Prior to joining the Oxford Robotics Institute, Oiwi was a DPhil in NLP (Oxford) and a Postdoc in Imaging Neuroscience (UCL and Oxford). At A2I, his research is focused on data-efficient Deep Learning and embodied Artificial Intelligence. One aim of this research is to draw inspiration from what we know about natural intelligence in biological brains to pioneer new methods in machine learning, particularly for use in agents that interact with the world. The desire for data-efficient Deep Learning (i.e. “small data, big computing”) is common to many applications in Clinical Neuroscience, Speech and Language Processing, and Engineering. Oiwi’s long-term goal is to develop Neural Speech Prosthetics for paralysed patients.
In his free time, Oiwi publishes on Polynesian languages.
Recent Publications
Reaching Through Latent Space: From Joint Statistics to Path Planning in Manipulation
Hung C-M, Zhong S, Goodwin W, Jones OP, Engelcke M et al. (2022)
BibTeX
@misc{reachingthrough-2022/10,
title={Reaching Through Latent Space: From Joint Statistics to Path Planning in
Manipulation},
author={Hung C-M, Zhong S, Goodwin W, Jones OP, Engelcke M et al.},
year = "2022"
}
Composing RNNs and FSTs for Small Data: Recovering Missing Characters in Old Hawaiian Text
Jones OP & Shillingford B (2022)
BibTeX
@misc{composingrnnsan-2022/7,
title={Composing RNNs and FSTs for Small Data: Recovering Missing Characters in
Old Hawaiian Text},
author={Jones OP & Shillingford B},
year = "2022"
}
ObPose: Leveraging Pose for Object-Centric Scene Inference in 3D
Wu Y, Jones OP & Posner I (2022)
BibTeX
@misc{obposeleveragin-2022/6,
title={ObPose: Leveraging Pose for Object-Centric Scene Inference in 3D},
author={Wu Y, Jones OP & Posner I},
year = "2022"
}
Generalizing Brain Decoding Across Subjects with Deep Learning
Csaky R, Es MV, Jones OP & Woolrich M (2022)
BibTeX
@misc{generalizingbra-2022/5,
title={Generalizing Brain Decoding Across Subjects with Deep Learning},
author={Csaky R, Es MV, Jones OP & Woolrich M},
year = "2022"
}
VAE-Loco: Versatile Quadruped Locomotion by Learning a Disentangled Gait Representation
Mitchell AL, Merkt W, Geisert M, Gangapurwala S, Engelcke M et al. (2022)
BibTeX
@misc{vaelocoversatil-2022/5,
title={VAE-Loco: Versatile Quadruped Locomotion by Learning a Disentangled Gait
Representation},
author={Mitchell AL, Merkt W, Geisert M, Gangapurwala S, Engelcke M et al.},
year = "2022"
}
Next Steps: Learning a Disentangled Gait Representation for Versatile Quadruped Locomotion
Mitchell AL, Merkt W, Geisert M, Gangapurwala S, Engelcke M et al. (2021)
BibTeX
@misc{nextstepslearni-2021/12,
title={Next Steps: Learning a Disentangled Gait Representation for Versatile
Quadruped Locomotion},
author={Mitchell AL, Merkt W, Geisert M, Gangapurwala S, Engelcke M et al.},
year = "2021"
}
APEX: Unsupervised, Object-Centric Scene Segmentation and Tracking for Robot Manipulation
Wu Y, Jones OP, Engelcke M & Posner I (2021)
BibTeX
@misc{apexunsupervise-2021/5,
title={APEX: Unsupervised, Object-Centric Scene Segmentation and Tracking for
Robot Manipulation},
author={Wu Y, Jones OP, Engelcke M & Posner I},
year = "2021"
}
GENESIS-V2: Inferring Unordered Object Representations without Iterative Refinement
Engelcke M, Jones OP & Posner I (2021)
BibTeX
@misc{genesisvinferri-2021/4,
title={GENESIS-V2: Inferring Unordered Object Representations without Iterative
Refinement},
author={Engelcke M, Jones OP & Posner I},
year = "2021"
}
There and Back Again: Learning to Simulate Radar Data for Real-World Applications
Weston R, Jones OP & Posner I (2020)
BibTeX
@misc{thereandbackaga-2020/11,
title={There and Back Again: Learning to Simulate Radar Data for Real-World
Applications},
author={Weston R, Jones OP & Posner I},
year = "2020"
}
Reconstruction Bottlenecks in Object-Centric Generative Models
Engelcke M, Jones OP & Posner I (2020)
BibTeX
@misc{reconstructionb-2020/7,
title={Reconstruction Bottlenecks in Object-Centric Generative Models},
author={Engelcke M, Jones OP & Posner I},
year = "2020"
}
