Enabling Graph Neural Networks at Exascale

Hvor vil det være trafikkork i morgen? Hvilke bivirkninger vil du få av medisinen? Hvor kommer falske nyheter fra, og hvordan spres de? Mange avgjørende spørsmål krever prediksjoner i en tilkoblet verden, og Graf Nervale Nettverk («GNN») lar oss komme med slike prediksjoner. Mens nåværende AI-systemer kan tolke bilder, for eksempel stoppskilt for selvkjørende biler, kan GNNer komme med forutsigelser om den fysiske verden, for eksempel biler som kommer til et veikryss, selv om de ikke kan sees for øyeblikket. I tilleg, GNNer kan brukes for a finner bedre løsninger til abstrakte problemer som kan gir bedre resultater i industrielle anvendelser. Et annet tema av spesiell interesse er analyse av sosiale nettverker da GNNer kan analysere og lære av nettverksforbindelser og tekstmeldinger samtidig. Siden spørsmålene vi ønsker å svare med GNNer er mer omfattende enn å gjenkjenne bilder, er det nødvendig med kraftigere datamaskiner for å betjene dem. Imidlertid er de kraftigste datamaskinene i dag sammensatt av mange mindre enheter, og bare ved å sikre at alle disse enhetene fungerer perfekt kan vi bruke datamaskiner effektivt. Dette er avgjørende for å utvide omfanget av hva GNN?er kan forutsi, og for å begrense energiforbruket til disse systemene.

-

Graph neural networks (GNNs), which extend the successful ideas of deep learning to irregularly structured data, are a recent addition to the field of artificial intelligence. While traditional deep learning has focused on regular inputs such as images composed of pixels in two-dimensional space, graph neural networks can analyze and learn from unstructured connections between objects. This gives GNNs the ability to tackle completely new classes of problems, such as analyzing social networks and power grids, or uncovering molecule structures in computational chemistry. Some experts in the field also believe that graph networks, due to their capacity for combinatorial generalization, represent an important next step towards the development of general artificial intelligence. However, such tasks require vast amounts of computation, which can only be provided by parallel processing. Is well known that parallel computation for irregular problems is much more challenging than for regular ones, and GNNs are no exception. While traditional deep learning has been scaled up to run on entire supercomputers efficiently, GNNs currently do not scale to multiple processors. This proposal aims to overcome this limitation by drawing upon decades of experience in scalable graph algorithms and sparse linear algebra, and adapting techniques that are proven to be effective for distributing graph computations over large parallel systems to GNNs. We aim to create a new computational framework which allows users to specify a GNN while the framework handles the task of distributing graphs over parallel machines, as well as selecting and running the algorithms that are best suited for the computation automatically. Recently, frameworks such as TensorFlow, have made traditional deep neural networks accessible for a large number of users. In the same way, our goal is to create a proof-of-concept framework that will be a crucial factor to the successful GNNs real-world appliance.