In this community review report, we discuss applications and techniques for fast machine learning (ML) in science—the concept of integrating powerful ML methods into the real-time experimental data processing loop to accelerate scientific discovery. The material for the report builds on two workshops held by the Fast ML for Science community and covers three main areas: applications for fast ML across a number of scientific domains; techniques for training and implementing performant and resource-efficient ML algorithms; and computing architectures, platforms, and technologies for deploying these algorithms. We also present overlapping challenges across the multiple scientific domains where common solutions can be found. This community report is intended to give plenty of examples and inspiration for scientific discovery through integrated and accelerated ML solutions. This is followed by a high-level overview and organization of technical advances, including an abundance of pointers to source material, which can enable these breakthroughs.

As neural networks have become more powerful, there has been a rising desire to deploy them in the real world; however, the power and accuracy of neural networks is largely due to their depth and complexity, making them difficult to deploy, especially in resource-constrained devices. Neural network quantization has recently arisen to meet this demand of reducing the size and complexity of neural networks by reducing the precision of a network. With smaller and simpler networks, it becomes possible to run neural networks within the constraints of their target hardware. This paper surveys the many neural network quantization techniques that have been developed in the last decade. Based on this survey and comparison of neural network quantization techniques, we propose future directions of research in the area.

The 1 l-cis to 1 1-trans torsional isomerization of retinal chromophore in rhodopsin has long been known to be involved in the early photochemistry of visual process [l]. Recent work on generation of femtosecond pulses in blue-green region of spectrum made it possible to study this process directly. We have done time-resolved absorption experiments on rhodopsin in both protonated and deuterated aqueous environments at room temperature [2,3]. These measurements test both the standard picture of rapid photoisomerization and also address the issue of whether proton translocation is also important in the initial step of vision [4]. A 500 nm 150 fs pump pulse initiates the reaction and a white light continuum probe is used to monitor absorbance transients in the 500 to 640 nm range. We observed two distinct kinetic components having 200 fs and 3 ps lifetimes. These data are well modeled by rate equations based on scheme depicted in Figure 1 which illustrates isomerization along the torsional coordinate of the ll-cis bond of the retinal chromophore. As illustrated, the 200 fs time can be associated with the appearance of a 90 degree twisted metastable intermediate by rotation around the double bond on the excited state surface and the 3 ps time with the decay of that intermediate to form the fully isomerized all-trans photoproduct known as bathorhodopsin. These times as well as the absorption spectrum (Figure 3) of the twisted metastable configuration agree well with the semiempirical energy level and molecular dynamics calculations of Tallent et a1. [5].Our data and recent model of initial photochemistry of vision conflict with the model proposed by Schoenlein et a1 [6] which postulate complete isomerization in 200 fs. The observed dynamics are insensitive to deuteration of the exchangeable protons which suggest that proton translocation is unimportant at physiological temperatures.