No abstract

No abstract

Deep Reinforcement Learning (DRL) is an avenue of research in Artificial Intelligence (AI) that has received increasing attention within the research community in recent years, and is beginning to show potential for real-world application. DRL is one of the most promising routes towards developing more autonomous AI systems that interact with and take actions in complex real-world environments, and can more flexibly solve a range of problems for which we may not be able to precisely specify a correct ‘answer’. This could have substantial implications for people’s lives: for example by speeding up automation in various sectors, changing the nature and potential harms of online influence, or introducing new safety risks in physical infrastructure. In this paper, we review recent progress in DRL, discuss how this may introduce novel and pressing issues for society, ethics, and governance, and highlight important avenues for future research to better understand DRL’s societal implications. This article appears in the special track on AI and Society.

Deep neural networks and decision trees operate on largely separate paradigms; typically, the former performs representation learning with prespecified architectures, while the latter is characterised by learning hierarchies over pre-specified features with data-driven architectures. We unite the two via adaptive neural trees (ANTs) that incorporates representation learning into edges, routing functions and leaf nodes of a decision tree, along with a backpropagation-based training algorithm that adaptively grows the architecture from primitive modules (e.g., convolutional layers). We demonstrate that, whilst achieving competitive performance on classification and regression datasets, ANTs benefit from (i) lightweight inference via conditional computation, (ii) hierarchical separation of features useful to the task e.g. learning meaningful class associations, such as separating natural vs. man-made objects, and (iii) a mechanism to adapt the architecture to the size and complexity of the training dataset.

Atopic dermatitis (AD), also known as eczema, is one of the most common chronic skin diseases. AD severity is primarily evaluated based on visual inspections by clinicians, but is subjective and has large inter-and intra-observer variability in many clinical study settings. To aid the standardisation and automating the evaluation of AD severity, this paper introduces a CNN computer vision pipeline, EczemaNet, that first detects areas of AD from photographs and then makes probabilistic predictions on the severity of the disease. EczemaNet combines transfer and multitask learning, ordinal classification, and ensembling over crops to make its final predictions. We test EczemaNet using a set of images acquired in a published clinical trial, and demonstrate low RMSE with well-calibrated prediction intervals. We show the effectiveness of using CNNs for non-neoplastic dermatological diseases with a medium-size dataset, and their potential for more efficiently and objectively evaluating AD severity, which has greater clinical relevance than mere classification.