Articles in 2022

Predicting the performance of a tactile sensor from its composition and morphology is nearly impossible with traditional computational approaches. Machine learning can not only predict device-level performance, but also recommend new material compositions for soft machine applications.

An international security conference explored how artificial intelligence (AI) technologies for drug discovery could be misused for de novo design of biochemical weapons. A thought experiment evolved into a computational proof.

Molecular representations are hard to design due to the large size of the chemical space, the amount of potentially important information in a molecular structure and the relatively low number of annotated molecules. Still, the quality of these representations is vital for computational models trying to predict molecular properties. Wang et al. present a contrastive learning approach to provide differentiable representations from unlabelled data.

Drug resistance in tropical diseases such as malaria requires constant improvement and development of new drugs. To find potential candidates, generative machine learning methods that can search for novel bioactive molecules can be employed.

Autonomous vehicle technologies need to be safer than humans by a considerable margin before they can be truly self-driving. But they can provide substantial benefits as assistive driving technology already today — provided their limitations are properly communicated.

Current AI policy recommendations differ on what the risks to human autonomy are. To systematically address risks to autonomy, we need to confront the complexity of the concept itself and adapt governance solutions accordingly.

Machine learning applications in agriculture can bring many benefits in crop management and productivity. However, to avoid harmful effects of a new round of technological modernization, fuelled by AI, a thorough risk assessment is required, to review and mitigate risks such as unintended socio-ecological consequences and security concerns associated with applying machine learning models at scale.

Machine learning performs well at predictive modelling based on statistical correlations, but for high-stakes applications, more robust, explainable and fair approaches are required. Cui and Athey discuss the benefits of bringing causal inference into machine learning, presenting a stable learning approach.

High-throughput single-cell sequencing data can provide valuable biological insights but are computationally challenging to analyse due to the dimensionality of the data and batch-specific biases. Kopp and colleagues have developed a variational auto-encoder-based method using a novel loss function for simultaneous batch correction and dimensionality reduction.

Tropical diseases, such as malaria, can develop resistance to specific drugs. Godinez and colleagues present here a generative design approach to find new anti-malarial drugs to circumvent this resistance.

The Large Hadron Collider produces 40 million collision events per second, most of which need to be discarded by a real-time filtering system. Unsupervised deep learning algorithms are developed and deployed on custom electronics to search for rare events indicating new physics, rather than for specific events led by theory.

High-fidelity haptic sensors with three-dimensional sensing surfaces are needed to advance dexterous robotic manipulation. The authors develop a sensor design that offers accurate force sensation across a three-dimensional surface while being robust, low-cost and easy to fabricate.

The combination of object recognition and viewpoint estimation is essential for visual understanding. However, convolutional neural networks often fail to generalize to object category–viewpoint combinations that were not seen during training. The authors investigate the impact of data diversity and architectural choices on the capability of generalizing to out-of-distribution combinations.

Controllers for robotic locomotion patterns deal with complex interactions and need to be carefully designed or extensively trained. Thor and Manoonpong present a modular approach for neural pattern generators that allows incremental and fast learning.

The investigation of single-cell epigenomics with technologies such as single-cell chromatin accessibility sequencing (scCAS) presents an opportunity to expand the understanding of gene regulation at the cellular level. The authors develop a probabilistic generative model to better characterize cell heterogeneity and accurately annotate the cell type of scCAS data.

Molecules are often represented as topological graphs while their true three-dimensional geometry contains a lot of valuable information. Xiaomin Fang and colleagues present a self-supervised molecule representation method that uses this geometric data in graph neural networks to predict a range of molecular properties.

Piezoresistors can be used in strain sensors for soft machines, but the traditional design process relies on intuition and human ingenuity alone. Haitao Yang and colleagues present a method built on genetic algorithms and other machine learning methods to design and fabricate strain sensors with improved capabilities.

Growing criticisms of datasets that were built from user-generated data scraped from the web have led to the retirement or redaction of many popular benchmarks. Their afterlife, as copies or subsets that continue to be used, is a cause for concern.