Stencil computations are widely used to simulate the change of state of physical systems. The current best algorithm for performing aperiodic linear stencil computations on a d (≥ 1)-dimensional grid of size N for T timesteps does Θ(TN1-1/d+N Log N) work. We introduce novel techniques based on random walks and Gaussian approximations for an asymptotic improvement of this work bound for a class of linear stencils. We also improve the span (i.e., parallel running time on an unbounded number of processors) asymptotically from the current state of the art.

Brief Announcement: Faster Stencil Computations using Gaussian Approximations

We study the binomial option pricing model and the Black-Scholes-Merton pricing model. In the binomial option pricing model, we concentrate on two widely-used call options: (1) European and (2) American. Under the Black-Scholes-Merton model, we investigate pricing American put options. Our contributions are two-fold: First, we transform the option pricing problems into nonlinear stencil computation problems and present efficient algorithms to solve them. Second, using our new FFT-based nonlinear stencil algorithms, we improve the work and span asymptotically for the option pricing problems we consider. In particular, we perform $O(T\log^2 T)$ work for both American call and put option pricing, where $T$ is the number of time steps.

Fast Option Pricing using Nonlinear Stencils

A Doctor’s Handwriting Recognition model can predict (recognize) the text present in the doctor’s prescription, by feeding image of that medicine name to the model, and it predicts the text present in the image and it gives the final medicine name as digital text. This model is suitable only for text written in English language and not suitable for other languages of texts written in prescription. The model based on training dataset the output it produces may get varied and based on training images count. Both convolution layers and Bi-Directional LSTM layers can be used for feature extraction and recognizing text, respectively. 

Interpreting Doctor’s Handwritten Prescription Using Deep Learning Techniques

Stencil computations are ubiquitous in modern grid-based physical simulations. In this paper, we present FOURST – a compiler to generate programs computing time iterated linear periodic and aperiodic stencil computations with fast Fourier transform methods. This paper outlines the design and implementation of the code generation approach in FOURST, to automatically generate FFT-based stencil solvers. We present experimental results on the state-of-the-art Ookami supercomputer housing Fujitsu A64FX and Intel Skylake processors, to study the performance of FOURST and a state-of-the-art tiling-based optimized code generator PLuTo on various stencil shapes and varying the number of time iterations. We discuss the performance profiles, and limitations, of both approaches on high-end modern hardware.

FOURST: A code generator for FFT-based fast stencil computations

The central dogma of handwritten character recognition remains inextricably linked to optical character recognition methods for print media. Alongside their reliance on proprietary data and lack of open-access software, the applicability of these optical character recognition methods to handwritten characters from low-quality documents (e.g., that are damaged) remains unknown. In this paper, we compare and contrast the performance of state-of-the-art optical character recognition tools for print and learning models engineered with state-of-the-art machine learning toolkits trained on handwritten inputs. Using Tesseract OCR as a baseline, we build, optimize, and evaluate three types of convolutional neural networks that are trained on the AL-ALLand AL-PUBdatasets, a collection of images of handwritten ancient Greek characters that were labeled by volunteers through the Ancient Lives online citizen science project. We find our best-performing machine learning model to be 92.57% accurate compared to Tesseract OCR’s 11.15%. Following our analysis, we present a brief examination of our models’ shortcomings, introduce the publicly-available AL-PUBdataset, and, describe Theia, a web-based tool that democratizes our machine learning models for public use. We conclude by discussing the promise of our findings for advancing research at the intersection of machine learning, manuscript transcription, and the digital humanities.

Gregory Croisdale

Papers

FOURST: A code generator for FFT-based fast stencil computations

Exploring Learning Approaches for Ancient Greek Character Recognition with Citizen Science Data