ORIGAMI accelerates ML algorithms by splitting compute tasks between in-memory and off-chip accelerators.
problem Memory bandwidth bottleneck in ML processing.
method Heterogeneous in-memory accelerators and off-chip compute platform, pattern-matching for compute patterns, computation-splitting compiler.
result ORIGAMI outperforms state-of-the-art accelerators in performance and energy-efficiency.
New algorithm trains binary-activation, multi-level RNNs for noise-resilient, ADC-/DAC-free PIM inference.
problem Training noise-resilient, ADC-/DAC-free neural networks.
method Binary activations and multi-level weights for eNVM-based processing-in-memory circuits.
result Higher accuracy and noise resilience for recurrent networks compared to existing methods.
Innovative neural networks reduce memory usage for efficient, accurate segmentation.
problem Efficiently segmenting large graphs with limited memory.
method Iterative neural networks with loops and multiple outputs.
result State-of-the-art semantic segmentation results on demanding datasets.
Unified framework for faster neural network training with less information loss.
problem Time-consuming backpropagation and loss of unpropagated gradient information.
method Unified sparse backpropagation framework and memorized sparse backpropagation algorithm.
result Convergence in probability with certain conditions and effective information loss mitigation.
Special issue honors Stanley Deser, focusing on advanced physics topics.
problem None explicitly stated in the abstract.
method Collection of articles in memory of Stanley Deser.
result No specific key result mentioned in the abstract.
Improved accuracy in neural networks using CTF memory with RPU.
problem Challenges in achieving high accuracy in neural networks using in-memory computing.
method Exploring the trade-off between conductance change range and linearity in CTF memory for RPU-based matrix operations.
result Achieved accuracy of 97.9% on MNIST dataset, 89.1% and 70.5% on CIFAR-10 and CIFAR-100 datasets.
Improved neural network training in low-dimensional random bases.
problem Inefficient optimization in large-scale neural networks.
method Re-draw random subspace at each training step, apply independent projections to different network parts.
result Significantly better optimization performance and efficiency.
We introduce a DNN training technique that learns only a fraction of the full parameter set without incurring an accuracy penalty. To do this, our algorithm constrains the total number of weights updated during backpropagation to those with the highest total gradients. The remaining weights are not tracked, and their i…
This research bridges binary and spiking neural networks for efficient on-chip AI.
problem Reducing compute requirements in machine learning frameworks.
method Training Spiking Neural Networks in extreme quantization regime and utilizing standard training techniques for conversion.
result Training Spiking Neural Networks in extreme quantization regime achieves near full precision accuracies.
Invites geometers to Garside theory for mapping class groups.
problem None explicitly stated, but related to geometric group theory.
method Garside theory applied to mapping class groups.
result No specific key result mentioned in the abstract.
Improved EXACT strategy reduces GNN memory consumption and runtime.
problem Efficiently training large-scale GNNs with reduced memory usage.
method Block-wise quantization of intermediate activation maps with improved variance minimization.
result Further reduction in memory consumption (>15%) and runtime speedup (5%) with similar performance trade-offs.
AIDEL improves scalability of learned indexes in storage systems.
problem Expensive retraining and heavy inter-model dependency in learned indexes limit scalability.
method Construct different linear regression models based on data distribution, making them independent and easier to partition.
result AIDEL improves insertion performance by about 2x and comparable lookup performance.
Trojans can be inserted into deep neural networks without altering training data.
problem Inserting malicious code into deep learning models without changing training data.
method Patches model parameters in memory to achieve desired behavior on specific inputs.
result Efficient patches can induce trojan behavior with minimal risk of detection.
RIn-Close_CVC is an efficient (take polynomial time per bicluster), complete (find all maximal biclusters), correct (all biclusters attend the user-defined level of consistency) and non-redundant (all the obtained biclusters are maximal and the same bicluster is not enumerated more than once) enumerative algorithm for …
New framework improves robustness of implicit neural networks.
problem Ill-posedness and convergence instability in implicit neural networks.
method NEMON framework based on contraction theory for ℓ∞ norm, including well-posedness condition, average iteration, and input-output Lipschitz constant regularization. result Improved accuracy and robustness of implicit models with smaller input-output Lipschitz bounds.
Automated framework optimizes DNN deployment on Arm CPUs.
problem Lack of globally optimised DNN deployment across software levels.
method Reinforcement Learning search for automated design space exploration.
result Up to 4x improvement in performance and 2x reduction in memory.
Rewards are sparse in the real world and most of today's reinforcement learning algorithms struggle with such sparsity. One solution to this problem is to allow the agent to create rewards for itself - thus making rewards dense and more suitable for learning. In particular, inspired by curious behaviour in animals, obs…
NSMs use always-on stochasticity to normalize activations, improving convergence and performance.
problem Improving the robustness and generalizability of deep neural networks.
method Developed Neural Sampling Machines (NSMs) using always-on multiplicative stochasticity and simple threshold neurons.
result NSMs exhibit self-normalizing properties similar to Weight Normalization, speeding up convergence and preventing internal covariate shift.
Natural spatiotemporal processes can be highly non-stationary in many ways, e.g. the low-level non-stationarity such as spatial correlations or temporal dependencies of local pixel values; and the high-level variations such as the accumulation, deformation or dissipation of radar echoes in precipitation forecasting. Fr…
Idioms pose problems to almost all Machine Translation systems. This type of language is very frequent in day-to-day language use and cannot be simply ignored. The recent interest in memory augmented models in the field of Language Modelling has aided the systems to achieve good results by bridging long-distance depend…
Accelerates Riemannian gradient methods with extrapolation.
problem Optimizing functions on manifolds efficiently.
method Extrapolating iterates in Riemannian gradient descent.
result Achieves optimal convergence rate and computational advantage.
Maximal acceleration metrics limit spacetime curvature.
problem Bounding spacetime curvature under maximal acceleration.
method Developed a geometric framework for maximal acceleration metrics and associated connections, proving curvature bounds.
result Uniform bounds on curvature components follow from uniform bounds on maximal acceleration.
Accelerates Q-learning by updating targets more efficiently.
problem Improving convergence speed in Q-learning algorithms.
method Proposes an accelerated target update scheme based on momentum methods.
result The proposed algorithms converge faster than vanilla Q-learning.
Accelerates optimization in asynchronous systems with sparse updates.
problem Optimizing finite-sum objectives in asynchronous lock-free environments.
method New accelerated SVRG variant with sparse updates.
result Achieves optimal incremental gradient complexity.
Service-induced congestion in memory-constrained LLM serving
problem Service-induced congestion in memory-constrained large language model (LLM) serving
method Developing a discrete-time dynamical model of memory-constrained LLM inference
result The system converges to a unique worst-case limit cycle that is asymptotically stable outside a Lebesgue-measure-zero exact-capture set, with throughput losses as large as 50%.
We analyze Riemannian accelerated methods using a new framework.
problem Understanding Riemannian accelerated gradient methods.
method Riemannian A-HPE framework, focusing on Euclidean A-HPE insights and metric distortion control.
result Characterization of acceleration for various Riemannian methods.
Super-acceleration of gradient descent with momentum improves loss function minimization.
problem Minimizing loss functions in machine learning.
method Extending Nesterov acceleration by using gradients at multiple steps ahead.
result Super-acceleration of the momentum algorithm is beneficial for various loss landscapes and tasks.
We describe novel subgradient methods for a broad class of matrix optimization problems involving nuclear norm regularization. Unlike existing approaches, our method executes very cheap iterations by combining low-rank stochastic subgradients with efficient incremental SVD updates, made possible by highly optimized and…
PF-LaCG removes the need for knowing smoothness and strong convexity parameters for locally accelerated CG.
problem Locally accelerated CG requires knowledge of smoothness and strong convexity parameters.
method Parameter-Free Locally Accelerated CG (PF-LaCG) algorithm.
result PF-LaCG achieves local acceleration without requiring knowledge of smoothness and strong convexity parameters.
Incremental versions of batch algorithms are often desired, for increased time efficiency in the streaming data setting, or increased memory efficiency in general. In this paper we present a novel algorithm for incremental kernel PCA, based on rank one updates to the eigendecomposition of the kernel matrix, which is mo…
Accelerates coordinate descent methods for machine learning problems.
problem Slowness of coordinate descent methods in machine learning.
method Extrapolation-based accelerated coordinate descent.
result Significant speed-up in practice compared to existing methods.
Acceleration in Hilbert spaces reduces computations but not accuracy.
problem Improving learning accuracy with fewer computations.
method Analysis of Nesterov acceleration and heavy-ball methods in Hilbert spaces.
result Acceleration can reduce computations but not improve accuracy with respect to gradient descent.
Continuized Nesterov acceleration accelerates stochastic gradient descent and gossip algorithms.
problem Improving the convergence rate of stochastic gradient descent and gossip algorithms.
method Introducing a continuized variant of Nesterov acceleration, which mixes variables continuously and takes gradient steps at random times.
result The continuized Nesterov acceleration achieves convergence rates similar to Nesterov's original acceleration but with random parameters.
Accelerated gradient methods play a central role in optimization, achieving optimal rates in many settings. While many generalizations and extensions of Nesterov's original acceleration method have been proposed, it is not yet clear what is the natural scope of the acceleration concept. In this paper, we study accelera…
Develops accelerated methods for optimization using low-dimensional projected-gradient information.
problem Optimization with low-dimensional projected-gradient information and Nesterov acceleration.
method Randomized-subspace Nesterov accelerated gradient methods for smooth convex and strongly convex optimization.
result Established accelerated oracle-complexity guarantees and unified basis for comparing sketch families.
Language Models (LMs) are important components in several Natural Language Processing systems. Recurrent Neural Network LMs composed of LSTM units, especially those augmented with an external memory, have achieved state-of-the-art results. However, these models still struggle to process long sequences which are more li…
Accelerates sampling from Gibbs distributions using ARWP method.
problem Sampling from Gibbs distributions efficiently.
method ARWP method, combining Nesterov acceleration and regularized Wasserstein proximal.
result ARWP exhibits higher contraction rate and faster tail exploration.
FedAc accelerates Federated Averaging for distributed optimization.
problem Efficiently optimizing distributed machine learning models.
method Federated Accelerated Stochastic Gradient Descent (FedAc) using a potential-based perturbed iterate analysis.
result FedAc achieves faster convergence and lower communication costs than previous methods.
This research accelerates sampling methods using Nesterov's Acceleration.
problem Improving sampling efficiency in MCMC methods.
method Developed a Hessian-Free High-Resolution ODE reformulation of NAG-SC, injected noise, and discretized the diffusion process.
result Quantified acceleration beyond underdamped Langevin in W2 distance for log-strongly-concave targets. Locally Accelerated Conditional Gradients improve convergence rates for smooth convex optimization problems.
problem Achieving optimal convergence rates for smooth convex optimization problems over polytopes.
method Locally Accelerated Conditional Gradients, coupling accelerated steps with conditional gradient steps.
result Achieves optimal accelerated local convergence for smooth strongly convex problems.
In this study, the concept of dual Lorentzian homotetic exponential motions in is discussed and their velocities, accelerations obtained. Also, some geometric results between velocity and acceleration vectors of a point in a spatial motion are obtained. Finally, the theorems related to acceleration and acceleration cen…
Improved analysis of accelerated noisy power method for PCA.
problem Inexact matrix-vector products in PCA settings.
method Improved analysis of Accelerated Noisy Power Method under milder perturbation conditions.
result Worst-case optimal convergence rate with relaxed noise conditions.
Variance reduction is a simple and effective technique that accelerates convex (or non-convex) stochastic optimization. Among existing variance reduction methods, SVRG and SAGA adopt unbiased gradient estimators and are the most popular variance reduction methods in recent years. Although various accelerated variants o…
New dynamical system framework explains Nesterov acceleration.
problem Understanding Nesterov's accelerated gradient method.
method Dynamical system derivation without vanishing step size.
result Acceleration arises from discretizing an ODE with semi-implicit Euler.
Hamiltonian dynamics-based algorithms achieve deterministic and accelerated convergence for convex optimization.
problem Accelerating convex optimization
method Hamiltonian dynamics
result Hamiltonian dynamics-based algorithms achieve deterministic and accelerated convergence for convex optimization.
AGBM accelerates GBM with theoretical guarantees.
problem Accumulation of errors in GBM's momentum term.
method Incorporates Nesterov's acceleration techniques and a corrected pseudo residual.
result First GBM type with theoretically-justified accelerated convergence rate.
New technique reduces memory usage and boosts neural network training speed.
problem Training large neural networks requires significant memory and computational resources.
method L2L (layer-to-layer) execution technique with eager param-server (EPS) and micro-batching.
result 45% reduction in memory usage and 40% increase in throughput for BERT-Large.
Accelerates MMLE using SVGD with Nesterov acceleration.
problem Maximum Marginal Likelihood Estimation optimization.
method Stein variational gradient descent with Nesterov acceleration.
result Consistently accelerates convergence across various tasks.