Jared Roesch
YOU?
Author Swipe
View article: Generalizing Shape Analysis with Gradual Types
Generalizing Shape Analysis with Gradual Types Open
Tensors are multi-dimensional data structures that can represent the data processed by machine learning tasks. Tensor programs tend to be short and readable, and they can leverage libraries and frameworks such as TensorFlow and PyTorch, as…
View article: Relax: Composable Abstractions for End-to-End Dynamic Machine Learning
Relax: Composable Abstractions for End-to-End Dynamic Machine Learning Open
Dynamic shape computations have become critical in modern machine learning workloads, especially in emerging large language models. The success of these models has driven the demand for their universal deployment across a diverse set of ba…
View article: Bring Your Own Codegen to Deep Learning Compiler
Bring Your Own Codegen to Deep Learning Compiler Open
Deep neural networks (DNNs) have been ubiquitously applied in many applications, and accelerators are emerged as an enabler to support the fast and efficient inference tasks of these applications. However, to achieve high model coverage wi…
View article: Safe functional systems through integrity types and verified assembly
Safe functional systems through integrity types and verified assembly Open
Building a trustworthy life-critical embedded system requires deep reasoning about the potential effects that sequences of machine instructions can have on full system operation. Rather than trying to analyze complete binaries and the coun…
View article: Dynamic Tensor Rematerialization
Dynamic Tensor Rematerialization Open
Checkpointing enables the training of deep learning models under restricted memory budgets by freeing intermediate activations from memory and recomputing them on demand. Current checkpointing techniques statically plan these recomputation…
View article: Nimble: Efficiently Compiling Dynamic Neural Networks for Model Inference
Nimble: Efficiently Compiling Dynamic Neural Networks for Model Inference Open
Modern deep neural networks increasingly make use of features such as dynamic control flow, data structures and dynamic tensor shapes. Existing deep learning systems focus on optimizing and executing static neural networks which assume a p…
View article: Theia: automatically generating correct program state visualizations
Theia: automatically generating correct program state visualizations Open
Program state visualizations (PSVs) help programmers understand hidden program state like objects, references, and closures. Unfortunately, existing PSV tools do not support custom language semantics, which educators often use to introduce…
View article: A Hardware–Software Blueprint for Flexible Deep Learning Specialization
A Hardware–Software Blueprint for Flexible Deep Learning Specialization Open
Specialized Deep Learning (DL) acceleration stacks, designed for a specific set of frameworks, model architectures, operators, and data types, offer the allure of high performance while sacrificing flexibility. Changes in algorithms, model…
View article: Relay: A High-Level IR for Deep Learning.
Relay: A High-Level IR for Deep Learning. Open
Frameworks for writing, compiling, and optimizing deep learning (DL) models have recently enabled progress in areas like computer vision and natural language processing. Extending these frameworks to accommodate the rapidly diversifying la…
View article: Relay: a new IR for machine learning frameworks
Relay: a new IR for machine learning frameworks Open
Machine learning powers diverse services in industry including search, translation, recommendation systems, and security. The scale and importance of these models require that they be efficient, expressive, and portable across an array of …
View article: A metaprogramming framework for formal verification
A metaprogramming framework for formal verification Open
We describe the metaprogramming framework currently used in Lean, an interactive theorem prover based on dependent type theory. This framework extends Lean's object language with an API to some of Lean's internal structures and procedures,…