Search

Effective connectivity (EC), indicative of the causal interactions between brain regions, is fundamental to understanding information processing in the brain. Traditional approaches, which infer EC from neural responses to stimulations, are not suited for mapping whole-brain EC in human due to being invasive and limited spatial coverage of stimulations. To address this gap, we present Neural Perturbational Inference (NPI), a data-driven framework designed to map EC across the entire brain. NPI employs an artificial neural network trained to learn large-scale neural dynamics as a computational surrogate of the brain. NPI maps EC by perturbing each region of the surrogate brain and observing the resulting responses in the rest of regions. NPI captures the directionality, strength, and excitatory/inhibitory properties of EC on a brain-wide scale. Our validation of NPI, using models with established EC, shows its superiority over Granger Causality and Dynamic Causal Modeling. Applying NPI to resting-state fMRI data from diverse datasets reveals consistent and structurally supported EC. Applications on a disease-specific dataset highlight the potential of using personalized EC as biomarkers for neurological diseases. By transitioning from correlational to causal understandings of brain functionality, NPI marks a stride in decoding the brain's functional architecture and can facilitate neuroscience research and clinical applications.

Recent studies show that deep learning models achieve good performance on medical imaging tasks such as diagnosis prediction. Among the models, multimodality has been an emerging trend, integrating different forms of data such as chest X-ray (CXR) images and electronic medical records (EMRs). However, most existing methods incorporate them in a model-free manner, which lacks theoretical support and ignores the intrinsic relations between different data sources. To address this problem, we propose a knowledge-driven and data-driven framework for lung disease diagnosis. By incorporating domain knowledge, machine learning models can reduce the dependence on labeled data and improve interpretability. We formulate diagnosis rules according to authoritative clinical medicine guidelines and learn the weights of rules from text data. Finally, a multimodal fusion consisting of text and image data is designed to infer the marginal probability of lung disease. We conduct experiments on a real-world dataset collected from a hospital. The results show that the proposed method outperforms the state-of-the-art multimodal baselines in terms of accuracy and interpretability.

In this paper, we provide a deep dive into the deployment of inference accelerators at Facebook. Many of our ML workloads have unique characteristics, such as sparse memory accesses, large model sizes, as well as high compute, memory and network bandwidth requirements. We co-designed a high-performance, energy-efficient inference accelerator platform based on these requirements. We describe the inference accelerator platform ecosystem we developed and deployed at Facebook: both hardware, through Open Compute Platform (OCP), and software framework and tooling, through Pytorch/Caffe2/Glow. A characteristic of this ecosystem from the start is its openness to enable a variety of AI accelerators from different vendors. This platform, with six low-power accelerator cards alongside a single-socket host CPU, allows us to serve models of high complexity that cannot be easily or efficiently run on CPUs. We describe various performance optimizations, at both platform and accelerator level, which enables this platform to serve production traffic at Facebook. We also share deployment challenges, lessons learned during performance optimization, as well as provide guidance for future inference hardware co-design.