Filter by:

  • With edge computing on custom hardware, real-time inference with deep neural networks can reach the nanosecond timescale. An important application in this regime is event processing at particle collision detectors like those at the Large Hadron Collider (LHC). To ensure high performance as well as reduced resource consumption, a method is developed, and made available as an extension of the Keras library, to automatically design optimal quantization of the different layers in a deep neural network.

    • Claudionor N. Coelho Jr
    • Aki Kuusela
    • Sioni Summers
  • Single-cell RNA sequencing efforts have made large amounts of data available for transcriptomics research. Simon and colleagues develop a neural network embedding approach that avoids batch effects, such that it can rapidly and efficiently integrate large datasets from different studies.

    • Lukas M. Simon
    • Yin-Ying Wang
    • Zhongming Zhao
  • Neural networks are becoming increasingly popular for applications in various domains, but in practice, further methods are necessary to make sure the models are learning patterns that agree with prior knowledge about the domain. A new approach introduces an explanation method, called ‘expected gradients’, that enables training with theoretically motivated feature attribution priors, to improve model performance on real-world tasks.

    • Gabriel Erion
    • Joseph D. Janizek
    • Su-In Lee
  • Monoclonalization, the isolation and expansion of a single cell derived from a cultured population, is an essential step in large-scale human cell culture and experiments. A new deep learning-based workflow called Monoqlo automatically detects colony presence and identifies clonality from cellular imaging, enabling single-cell selection protocols to be scalable while minimizing technical variability.

    • Brodie Fischbacher
    • Sarita Hedaya
    • Daniel Paull
  • The urgency of the developing COVID-19 epidemic has led to a large number of novel diagnostic approaches, many of which use machine learning. DeGrave and colleagues use explainable AI techniques to analyse a selection of these approaches and find that the methods frequently learn to identify features unrelated to the actual disease.

    • Alex J. DeGrave
    • Joseph D. Janizek
    • Su-In Lee
  • Gaining access to medical data to train AI applications can present problems due to patient privacy or proprietary interests. A way forward can be privacy-preserving federated learning schemes. Kaissis, Ziller and colleagues demonstrate here their open source framework for privacy-preserving medical image analysis in a remote inference scenario.

    • Georgios Kaissis
    • Alexander Ziller
    • Rickmer Braren
  • In the last few years, computational protein structure prediction has greatly advanced by combining deep learning including convolutional residual networks (ResNet) with co-evolution data. A new study finds that using deeper and wider ResNets improves predictions in the absence of co-evolution information, suggesting that the ResNets do not not simply de-noise co-evolution signals, but instead may learn important protein sequence–structure relationships.

    • Jinbo Xu
    • Matthew McPartlon
    • Jin Li
  • Calcium imaging is a valuable tool for recording in vivo neural activity, but the task of extracting signals of individual neurons is computationally challenging. Bao and colleagues present a U-Net-based method that is both accurate and fast enough to potentially allow real-time processing and closed-loop experiments.

    • Yijun Bao
    • Somayyeh Soltanian-Zadeh
    • Yiyang Gong
  • Experimental benchmarks such as ImageNet and Atari games play an important part in advancing artificial intelligence research. An analysis of results and papers linked to 25 popular benchmarks shows that research dynamics beyond conventional co-authorship has developed in this area.

    • Fernando Martínez-Plumed
    • Pablo Barredo
    • José Hernández-Orallo
  • The movement of drone swarms can be coordinated using virtual potential fields to reach a global goal and avoid local collisions. Soria et al. propose here to extend potential fields with a predictive model that takes into account the agents’ flight dynamics to improve the speed and safety of the swarm.

    • Enrica Soria
    • Fabrizio Schiano
    • Dario Floreano
  • Although technologies enable large-scale profiling of chromatin accessibility at the single-cell level, there are methodological challenges due to high dimensionality and high sparsity of data. Liu and colleagues describe a computational tool for the simultaneous determination of latent representation and clustering of cells from single-cell ATAC-seq data using a pair of generative adversarial networks.

    • Qiao Liu
    • Shengquan Chen
    • Wing Hung Wong
  • There is an urgent need to identify drugs that may be effective against SARS-CoV-2. A platform with a range of machine learning models is made available to predict anti-COVID-19 activity in candidate drugs and to help prioritize compounds for virtual screening.

    • Govinda B. KC
    • Giovanni Bocci
    • Tudor I. Oprea
  • Tackling scientific problems often requires computational models that bridge several spatial and temporal scales. A new simulation framework employing machine learning, which is scalable and can be used on standard laptops as well as supercomputers, promises exhaustive multiscale explorations.

    • Harsh Bhatia
    • Timothy S. Carpenter
    • Peer-Timo Bremer
  • Recurrent neural networks (RNNs) can learn to process temporal information, such as speech or movement. New work makes such approaches more powerful and flexible by describing theory and experiments demonstrating that RNNs can learn from a few examples to generalize and predict complex dynamics including chaotic behaviour.

    • Jason Z. Kim
    • Zhixin Lu
    • Danielle S. Bassett
  • Bats with sophisticated biosonar systems move their ears at a high speed to help localize sound sources. Yin and Müller present a system inspired by this strategy, which can localize sounds with high accuracy and with a single detector, using a flexible silicone model of a bat’s ear and a deep convolutional neural network to process the complex Doppler signatures.

    • Xiaoyan Yin
    • Rolf Müller
  • Predicting what comes next in a previously unseen time series of input data is a challenging task for machine learning. A novel unsupervised learning scheme termed predictive principal component analysis can extract the most informative components for predicting future inputs with low computational cost.

    • Takuya Isomura
    • Taro Toyoizumi
  • Identifying cancer driver genes from high-throughput genomic data is an important task to understand the molecular basis of cancer and to develop new treatments including precision medicine. To tackle this challenge, EMOGI, a new deep learning method based on graph convolutional networks is developed, which combines protein–protein interaction networks with multiomics datasets.

    • Roman Schulte-Sasse
    • Stefan Budach
    • Annalisa Marsico
  • Rechargeable lithium-ion batteries play a crucial role in many modern-day applications, including portable electronics and electric vehicles, but they degrade over time. To ensure safe operation, a battery’s ‘state of health’ should be monitored in real time, and this machine learning pipeline, tested on a variety of charging conditions, can provide such an online estimation of battery state of health.

    • Darius Roman
    • Saurabh Saxena
    • David Flynn
  • A generative approach called SliceGAN is demonstrated that can construct complex three-dimensional (3D) images from representative two-dimensional (2D) image examples. This is a promising approach in particular for studying microstructured materials where acquiring good-quality 3D data is challenging; 3D datasets can be created with SliceGAN, making use of high-quality 2D imaging techniques that are widely available.

    • Steve Kench
    • Samuel J. Cooper