Self-supervised learning allows training of neural networks without immense, high-quality or labeled datasets. We demonstrate that self-supervision furthermore improves robustness models using small, imbalanced incomplete datasets which pose severe difficulties to supervised models. For small datasets, the accuracy our approach is up 12.5% higher MNIST and 15.2% Fashion-MNIST compared random in...

Labeling training examples at scale is a perennial challenge in machine learning. Self-supervision methods compensate for the lack of direct supervision by leveraging prior knowledge to automatically generate noisy labeled examples. Deep probabilistic logic (DPL) unifying framework self-supervised learning that represents unknown labels as latent variables and incorporates diverse self-supervis...

Using consensual qualitative research, researchers interviewed 16 supervisors regarding their use of self-disclosure in supervision. Supervisors reported that their prior training in supervisor self-disclosure (SRSD) came via didactic sources and encouraged judicious use of SRSD. Supervisors used SRSD to enhance supervisee development and normalize their experiences; supervisors did not use SRS...

This study examines the relationship between level of supervision by the juvenile probation officers (JPO) and an adolescent's offending, considering the characteristics of juvenile offenders (specifically, level of psychopathy). Data are taken from the Pathways to Desistance Study on a subset of 859 juvenile offenders. We found that the level of probation officer supervision was not consistent...

This work aims at improving instance retrieval with self-supervision. We find that fine-tuning using the recently developed self-supervised learning (SSL) methods, such as SimCLR and MoCo, fails to improve performance of retrieval. In this work, we identify learnt representations for should be invariant large variations in viewpoint background etc., whereas self-augmented positives applied by c...

BACKGROUND This paper concerns a reflection on one interdisciplinary doctorate supervision project. It outlines key elements for success within this supervisory context. AIM The aim of this paper is to present a reflection on interdisciplinary doctorate supervision project experience and examine these experiences in light of literature on the topic. METHODS Reflection was carried out using ...

BACKGROUND Community health workers (CHWs) are an increasingly important component of health systems and programs. Despite the recognized role of supervision in ensuring CHWs are effective, supervision is often weak and under-supported. Little is known about what constitutes adequate supervision and how different supervision strategies influence performance, motivation, and retention. OBJECTI...