A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.
Paper proposes new gradient codes for robust distributed machine learning.
Gradient coding is a technique for straggler mitigation in distributed learning. In this paper we design novel gradient codes using tools from classical coding theory, namely, cyclic MDS codes, which compare favorably with existing solutions, both in the applicable range of parameters and in the complexity of the invol…
We present ErasureHead, a new approach for distributed gradient descent (GD) that mitigates system delays by employing approximate gradient coding. Gradient coded distributed GD uses redundancy to exactly recover the gradient at each iteration from a subset of compute nodes. ErasureHead instead uses approximate gradien…
Distributed implementations of gradient-based methods, wherein a server distributes gradient computations across worker machines, suffer from slow running machines, called 'stragglers'. Gradient coding is a coding-theoretic framework to mitigate stragglers by enabling the server to recover the gradient sum in the prese…
New gradient coding schemes reduce decoding error in both random and adversarial straggler settings.
Gradient codes adapt to varying straggler counts in distributed learning.
STRATA generates code adversarial examples efficiently without gradients.
Coded computation techniques provide robustness against straggling servers in distributed computing, with the following limitations: First, they increase decoding complexity. Second, they ignore computations carried out by straggling servers; and they are typically designed to recover the full gradient, and thus, canno…
Gradient descent and its many variants, including mini-batch stochastic gradient descent, form the algorithmic foundation of modern large-scale machine learning. Due to the size and scale of modern data, gradient computations are often distributed across multiple compute nodes. Unfortunately, such distributed implement…
Distributed algorithms are often beset by the straggler effect, where the slowest compute nodes in the system dictate the overall running time. Coding-theoretic techniques have been recently proposed to mitigate stragglers via algorithmic redundancy. Prior work in coded computation and gradient coding has mainly focuse…
We propose a novel coding theoretic framework for mitigating stragglers in distributed learning. We show how carefully replicating data blocks and coding across gradients can provide tolerance to failures and stragglers for Synchronous Gradient Descent. We implement our schemes in python (using MPI) to run on Amazon EC…
This paper develops coding techniques to reduce the running time of distributed learning tasks. It characterizes the fundamental tradeoff to compute gradients (and more generally vector summations) in terms of three parameters: computation load, straggler tolerance and communication cost. It further gives an explicit c…
This work improves federated learning privacy and accuracy with non-private data sharing and approximate gradient coding.
We use matricial free energy to regularize autoencoders, producing Gaussian-like codes.
This paper considers the problem of implementing large-scale gradient descent algorithms in a distributed computing setting in the presence of {\em straggling} processors. To mitigate the effect of the stragglers, it has been previously proposed to encode the data with an erasure-correcting code and decode at the maste…
Automatic differentiation (AD) is an essential primitive for machine learning programming systems. Tangent is a new library that performs AD using source code transformation (SCT) in Python. It takes numeric functions written in a syntactic subset of Python and NumPy as input, and generates new Python functions which c…
This paper studies communication efficiency in federated learning by optimizing the sum-rate-distortion function for indirect multiterminal source coding.
StarNet trains deep models without gradients using linear equations.
Bayesian Predictive Coding improves deep learning uncertainty quantification.
Study shows code-level optimizations significantly impact deep RL algorithms.
CodedFedL speeds up federated learning in MEC networks by 15x.
New hashing method improves document retrieval precision.
Optimizing over-the-air convex optimization, analog schemes are nearly optimal at low SNR.
Bio-inspired neural networks use predictive coding for efficient weight updates.
LoRA-One uses one-step full gradient to align adapters for efficient large model fine-tuning.
Distributed gradient descent (DGD) is an efficient way of implementing gradient descent (GD), especially for large data sets, by dividing the computation tasks into smaller subtasks and assigning to different computing servers (CSs) to be executed in parallel. In standard parallel execution, per-iteration waiting time …
GradPower speeds up language model training with minimal code changes.
In this work, we introduce a deep learning-based polar code construction algorithm. The core idea is to represent the information/frozen bit indices of a polar code as a binary vector which can be interpreted as trainable weights of a neural network (NN). For this, we demonstrate how this binary vector can be relaxed t…
The natural gradient method has been used effectively in conjugate Gaussian process models, but the non-conjugate case has been largely unexplored. We examine how natural gradients can be used in non-conjugate stochastic settings, together with hyperparameter learning. We conclude that the natural gradient can signific…
New method learns sparse distributions by thresholding samples, improving performance and efficiency.
New theory shows predictive coding makes learning landscape easier to navigate.
New method tackles high-dimensional SBL without covariance matrices.
Paper proposes LC-Checkpoint for efficient deep learning model checkpoints.
A simple continuous relaxation for argsort improves performance and is easy to implement.
This work improves dictionary learning speed without sacrificing accuracy.
We propose an efficient algorithm for the generalized sparse coding (SC) inference problem. The proposed framework applies to both the single dictionary setting, where each data point is represented as a sparse combination of the columns of one dictionary matrix, as well as the multiple dictionary setting as given in m…
Many continuous control tasks have bounded action spaces. When policy gradient methods are applied to such tasks, out-of-bound actions need to be clipped before execution, while policies are usually optimized as if the actions are not clipped. We propose a policy gradient estimator that exploits the knowledge of action…
Gradient descent optimization improved by circuit perspective.
This paper introduces Jensen, an easily extensible and scalable toolkit for production-level machine learning and convex optimization. Jensen implements a framework of convex (or loss) functions, convex optimization algorithms (including Gradient Descent, L-BFGS, Stochastic Gradient Descent, Conjugate Gradient, etc.), …
Modern large scale machine learning applications require stochastic optimization algorithms to be implemented on distributed computational architectures. A key bottleneck is the communication overhead for exchanging information such as stochastic gradients among different workers. In this paper, to reduce the communica…
PUDLE method analyzes and improves unrolled sparse coding networks for dictionary learning.
We study gradient compression methods to alleviate the communication bottleneck in data-parallel distributed optimization. Despite the significant attention received, current compression schemes either do not scale well or fail to achieve the target test accuracy. We propose a new low-rank gradient compressor based on …
Most optimizers including stochastic gradient descent (SGD) and its adaptive gradient derivatives face the same problem where an effective learning rate during the training is vastly different. A learning rate scheduling, mostly tuned by hand, is usually employed in practice. In this paper, we propose CProp, a gradient…
A novel method to propagate uncertainty through the soft-thresholding nonlinearity is proposed in this paper. At every layer the current distribution of the target vector is represented as a spike and slab distribution, which represents the probabilities of each variable being zero, or Gaussian-distributed. Using the p…
Embedding methods such as word embedding have become pillars for many applications containing discrete structures. Conventional embedding methods directly associate each symbol with a continuous embedding vector, which is equivalent to applying linear transformation based on "one-hot" encoding of the discrete symbols. …
In "Dictionary Learning" one tries to recover incoherent matrices (typically overcomplete and whose columns are assumed to be normalized) and sparse vectors with a small support of size for some while having access to observations $y \in \mathbb{…
New method fine-tunes discrete diffusion models for RLHF tasks.
Adversarial examples have recently proven to be able to fool deep learning methods by adding carefully crafted small perturbation to the input space image. In this paper, we study the possibility of generating adversarial examples for code-based iris recognition systems. Since generating adversarial examples requires b…