Related papers: The BLAS API of BLASFEO: optimizing performance fo…

Bilevel optimization has shown its utility across various machine learning settings, yet most algorithms in practice require second-order information, making it challenging to scale them up. Only recently, a paradigm of first-order…

The growing demand for long-context inference capabilities in Large Language Models (LLMs) has intensified the computational and memory bottlenecks inherent to the self-attention mechanism. To address this challenge, we introduce BLASST, a…

Large-scale deployment of generative AI tools often depends on costly API calls to a Large Language Model (LLM) to fulfil user queries. To curtail the frequency of these calls, one can employ a smaller language model -- a student -- which…

Limbo is an open-source C++11 library for Bayesian optimization which is designed to be both highly flexible and very fast. It can be used to optimize functions for which the gradient is unknown, evaluations are expensive, and runtime cost…

Computational implementations for solving systems of linear equations often rely on a one-size-fits-all approach based on LU decomposition of dense matrices stored in column-major format. Such solvers are typically implemented with the aid…

Sparse matrices are an integral part of scientific simulations. As hardware evolves new sparse matrix storage formats are proposed aiming to exploit optimizations specific to the new hardware. In the era of heterogeneous computing, users…

Global optimization of black-box functions is challenging in high dimensions. We introduce a conceptual adaptive random search framework, Branching Adaptive Surrogate Search Optimization (BASSO), that combines partitioning and surrogate…

The rapid development of RISC-V instruction set architecture presents new opportunities and challenges for software developers. Is it sufficient to simply recompile high-performance software optimized for x86-64 onto RISC-V CPUs? Are…

High-throughput inference serving is essential for applications built on large language models (LLMs). Existing serving frameworks reduce request-level and batch-level bubbles through batching and scheduling, but often overlook bubbles…

When handling large datasets that exceed the capacity of the main memory, movement of data between main memory and external memory (disk), rather than actual (CPU) computation time, is often the bottleneck in the computation. Since data is…

We introduce a library called Push that takes a probabilistic programming approach to Bayesian deep learning (BDL). This library enables concurrent execution of BDL inference algorithms on multi-GPU hardware for neural network (NN) models.…

Task-based programming models are excellent tools to parallelize and seamlessly load balance an application workload. However, the integration of I/O intensive applications and task-based programming models is lacking. Typically, I/O…

We propose PAIO, the first general-purpose framework that enables system designers to build custom-made Software-Defined Storage (SDS) data plane stages. It provides the means to implement storage optimizations adaptable to different…

Large language models (LLMs) have become increasingly popular in various areas, traditional business gradually shifting from rule-based systems to LLM-based solutions. However, the inference of LLMs is resource-intensive or…

Along with the progress of AI democratization, machine learning (ML) has been successfully applied to edge applications, such as smart phones and automated driving. Nowadays, more applications require ML on tiny devices with extremely…

We present POLO --- a C++ library for large-scale parallel optimization research that emphasizes ease-of-use, flexibility and efficiency in algorithm design. It uses multiple inheritance and template programming to decompose algorithms into…

Basic Linear Algebra Subprograms (BLAS) play key role in high performance and scientific computing applications. Experimentally, yesteryear multicore and General Purpose Graphics Processing Units (GPGPUs) are capable of achieving up to 15…

Numerical methods of the ADER family, in particular finite-element ADER-DG and finite-volume ADER-WENO methods, are among the most accurate numerical methods for solving quasilinear hyperbolic PDE systems. The internal structure of ADER-DG…

In recommendation systems, practitioners observed that increase in the number of embedding tables and their sizes often leads to significant improvement in model performances. Given this and the business importance of these models to major…

In edge computing, suppressing data size is a challenge for machine learning models that perform complex tasks such as autonomous driving, in which computational resources (speed, memory size and power) are limited. Efficient lossy…