arXiv research

This study investigates the use of reinforcement learning to guide a general purpose cache manager decisions. Cache managers directly impact the overall performance of computer systems. They govern decisions about which objects should be cached, the duration they should be cached for, and decides on which objects to ev…

A deep learning model predicts future cache accesses with high accuracy.

Parrot learns optimal cache replacement policies using imitation learning.

DEAP Cache learns prefetching, eviction, and admission using machine learning.

KVCOMM optimizes multi-agent LLM systems by reusing KV-caches, reducing redundant processing.

Support Vector Machines (SVMs) can solve structured multi-output learning problems such as multi-label classification, multiclass classification and vector regression. SVM training is expensive especially for large and high dimensional datasets. The bottleneck of the SVM training often lies in the kernel value computat…

We developed a caching method to speed up concept learning in complex knowledge bases.

Training large-scale image recognition models is computationally expensive. This raises the question of whether there might be simple ways to improve the test performance of an already trained model without having to re-train or fine-tune it with new data. Here, we show that, surprisingly, this is indeed possible. The …

New algorithm improves cache management with delayed feedback and decaying costs.

New method speeds up diffusion models without sacrificing quality.

Due to the drastic increase of mobile traffic, wireless caching is proposed to serve repeated requests for content download. To determine the caching scheme for decentralized caching networks, the content preference learning problem based on mobility prediction is studied. We first formulate preference prediction as a …

This paper optimizes caching and model multiplexing for large model inference.

A new method for energy-efficient file delivery in small cell networks.

Machine learning models that take computer program source code as input typically use Natural Language Processing (NLP) techniques. However, a major challenge is that code is written using an open, rapidly changing vocabulary due to, e.g., the coinage of new variable and method names. Reasoning over such a vocabulary i…

Over the last few years, Deep Neural Networks (DNNs) have become ubiquitous owing to their high accuracy on real-world tasks. However, this increase in accuracy comes at the cost of computationally expensive models leading to higher prediction latencies. Prior efforts to reduce this latency such as quantization, model …

Existing multi-armed bandit (MAB) models make two implicit assumptions: an arm generates a payoff only when it is played, and the agent observes every payoff that is generated. This paper introduces synchronization bandits, a MAB variant where all arms generate costs at all times, but the agent observes an arm's instan…

LATM framework uses LLMs to create and reuse tools for efficient problem-solving.

In this paper, the distributed edge caching problem in fog radio access networks (F-RANs) is investigated. By considering the unknown spatio-temporal content popularity and user preference, a user request model based on hidden Markov process is proposed to characterize the fluctuant spatio-temporal traffic demands in F…

Typical large-scale recommender systems use deep learning models that are stored on a large amount of DRAM. These models often rely on embeddings, which consume most of the required memory. We present Bandana, a storage system that reduces the DRAM footprint of embeddings, by using Non-volatile Memory (NVM) as the prim…

FibQuant improves KV-cache compression for long-context inference.

We address challenges of active learning under scarce informational resources in non-stationary environments. In real-world settings, data labeled and integrated into a predictive model may become invalid over time. However, the data can become informative again with switches in context and such changes may indicate un…

Research optimizes C++ patterns for HFT, reducing latency and improving profitability.

This paper is about metric data structures in high-dimensional or non-Euclidean space that permit cached sufficient statistics accelerations of learning algorithms. It has recently been shown that for less than about 10 dimensions, decorating kd-trees with additional "cached sufficient statistics" such as first and sec…

Low power digital signal processors (DSPs) typically have a very limited amount of memory in which to cache data. In this paper we develop efficient bottleneck feature (BNF) extractors that can be run on a DSP, and retrain a baseline large-vocabulary continuous speech recognition (LVCSR) system to use these BNFs with o…

Generating logical form equivalents of human language is a fresh way to employ neural architectures where long short-term memory effectively captures dependencies in both encoder and decoder units. The logical form of the sequence usually preserves information from the natural language side in the form of similar token…

In this paper, we propose several improvements on the block-coordinate Frank-Wolfe (BCFW) algorithm from Lacoste-Julien et al. (2013) recently used to optimize the structured support vector machine (SSVM) objective in the context of structured prediction, though it has wider applications. The key intuition behind our i…

In this paper we analyze, evaluate, and improve the performance of training generalized linear models on modern CPUs. We start with a state-of-the-art asynchronous parallel training algorithm, identify system-level performance bottlenecks, and apply optimizations that improve data parallelism, cache line locality, and …

Hyperparameter tuning of multi-stage pipelines introduces a significant computational burden. Motivated by the observation that work can be reused across pipelines if the intermediate computations are the same, we propose a pipeline-aware approach to hyperparameter tuning. Our approach optimizes both the design and exe…

Paper accelerates K-means clustering for large sparse document data.

Transformer is a popularly used neural network architecture, especially for language understanding. We introduce an extended and unified architecture that can be used for tasks involving a variety of modalities like image, text, videos, etc. We propose a spatio-temporal cache mechanism that enables learning spatial dim…

Developing efficient and scalable algorithms for Latent Dirichlet Allocation (LDA) is of wide interest for many applications. Previous work has developed an O(1) Metropolis-Hastings sampling method for each token. However, the performance is far from being optimal due to random accesses to the parameter matrices and fr…

Efficiently combines autoregressive and set-based models for joint distributions.

Paper details Hilbert-curve for high-performance data mining.

Automated process discovery is a class of process mining methods that allow analysts to extract business process models from event logs. Traditional process discovery methods extract process models from a snapshot of an event log stored in its entirety. In some scenarios, however, events keep coming with a high arrival…

Paper proposes EEIPU, a memoization-aware BO algorithm to reduce hyperparameter tuning costs.

Study uses three sources to evaluate language models fairly.

New algorithm reduces costs and latency for large language model inference.

In this paper, we propose a dual memory structure for reinforcement learning algorithms with replay memory. The dual memory consists of a main memory that stores various data and a cache memory that manages the data and trains the reinforcement learning agent efficiently. Experimental results show that the dual memory …

The adjoint sensitivity method scalably computes gradients of solutions to ordinary differential equations. We generalize this method to stochastic differential equations, allowing time-efficient and constant-memory computation of gradients with high-order adaptive solvers. Specifically, we derive a stochastic differen…

Service-induced congestion in memory-constrained LLM serving

Convolutional autoregressive models have recently demonstrated state-of-the-art performance on a number of generation tasks. While fast, parallel training methods have been crucial for their success, generation is typically implemented in a naïve fashion where redundant computations are unnecessarily repeated. This res…

We present CYCLADES, a general framework for parallelizing stochastic optimization algorithms in a shared memory setting. CYCLADES is asynchronous during shared model updates, and requires no memory locking mechanisms, similar to HOGWILD!-type algorithms. Unlike HOGWILD!, CYCLADES introduces no conflicts during the par…

DeltaGrad rapidly retrain models with minimal data changes.

DaRE forests enable efficient data deletion from random forests.

The paper proposes a method for interpretable mixture density estimation using a tree structure.

We develop BatchBALD, a tractable approximation to the mutual information between a batch of points and model parameters, which we use as an acquisition function to select multiple informative points jointly for the task of deep Bayesian active learning. BatchBALD is a greedy linear-time $1 - \frac{1}{e}$-approximate a…

The TensorFlow Distributions library implements a vision of probability theory adapted to the modern deep-learning paradigm of end-to-end differentiable computation. Building on two basic abstractions, it offers flexible building blocks for probabilistic computation. Distributions provide fast, numerically stable metho…

Optimizes web page freshness with limited crawling frequencies.