New layer improves neural networks' flexibility and speed.
problem Designing efficient neural networks for real-time learning.
method Sparse Associative Memories for incremental learning.
result Improved Neural Networks with faster training and real-time support.
Sparse neural encoding can store more memories as targets become sparser.
problem Storing sparse input-target associations in neural networks.
method Mathematical proofs using properties of random polytopes and sub-gaussian random vector variables.
result The capacity of neural maps increases with sparsity in target layers.
Sparse Hopfield model improves memory retrieval with fewer connections.
problem Memory retrieval efficiency with fewer connections.
method Sparse extension of Hopfield model, derived from sparse entropic regularizer.
result Sparse Hopfield model achieves tighter error bounds and better performance.
STanHop predicts multivariate time series with memory-enhanced capabilities.
problem Predicting multivariate time series with memory-enhanced capabilities.
method Sparse Tandem Hopfield Network (STanHop) with two external memory modules.
result STanHop outperforms dense Hopfield models in memory retrieval error.
Neural network uses local credit assignment to learn distant cause-effect relationships.
problem Learning distant cause-effect relationships in sequential data.
method Sparse coding in a recurrent neural network memory with local and immediate credit assignment.
result Network can predict partially-observable higher-order sequences and navigate mazes.
Improves sparse reinforcement learning efficiency with OYMB.
problem Sparse rewards hinder reinforcement learning performance.
method Introduces OYMB, a sampler for HER to control minibatch makeup.
result HER combined with OYMB leads to faster real goal completion.
New GPU kernels boost deep learning speed and memory efficiency.
problem Sparse deep learning matrices are not well-suited for existing sparse kernels.
method Identified favorable properties of sparse matrices from deep learning, developed high-performance GPU kernels for sparse matrix operations.
result 27% of single-precision peak performance on Nvidia V100 GPUs achieved with new kernels.
Global Memory Augmentation (GMAT) improves Transformer performance on long documents.
problem Large memory requirements of Transformer pairwise dot-product attention for long sequences.
method Integrates a dense global memory of length M into sparse Transformer blocks.
result Significant improvement on various tasks, including synthetic tasks, masked language modeling, and reading comprehension.
Guarantees sparse recovery for neural networks with iterative hard thresholding.
problem Recovering sparse network weights in neural networks.
method Structural properties of sparse network weights and iterative hard thresholding algorithm.
result Simple iterative hard thresholding algorithm recovers sparse network weights exactly using linear memory.
This paper proposes a new weight representation scheme for efficient model compression and performance enhancement.
problem Challenges in achieving performance enhancement on devices due to irregular sparse matrix representations.
method Fine-grained and unstructured pruning method combined with structured weight encryption.
result Achieved high compression ratios and performance on various deep learning models.
Deep neural networks (DNNs) have emerged as key enablers of machine learning. Applying larger DNNs to more diverse applications is an important challenge. The computations performed during DNN training and inference are dominated by operations on the weight matrices describing the DNN. As DNNs incorporate more layers a…
Model learns continuously from text without forgetting.
problem Catastrophic forgetting in lifelong language learning.
method Episodic memory with sparse experience replay and local adaptation.
result Model can continuously learn from new datasets.
Efficient sparse modern Hopfield models are introduced for memory retrieval and learning tasks.
problem Efficient modern Hopfield models for memory retrieval and learning tasks.
method Nonparametric interpretation of Hopfield models as a regression problem.
result Sparse-structured modern Hopfield models with sub-quadratic complexity.
Efficiently scales continuous kernels with sparse Fourier domain learning.
problem High computational and memory demands, spectral bias in continuous kernels.
method Sparse learning in the Fourier domain.
result Efficient scaling of continuous kernels, reduced computational and memory requirements, mitigated spectral bias.
Online method learns sparse models efficiently in large scale settings.
problem Sparse model learning in large scale settings with high computational and memory costs.
method Online learning approach, mini-batch methods, hard thresholding based stochastic gradient algorithm.
result Sparsity promoted by batch methods is not preserved in online fashion.
Sparse Transformers reduce memory and time requirements for long sequence modeling.
problem Quadratic growth in memory and time with transformer sequence length.
method Sparse factorizations of attention matrix, deeper network training, attention matrix recomputation, fast attention kernels.
result Sparse Transformers can model sequences up to tens of thousands of timesteps with hundreds of layers.
Proposes a novel SAM operator for separate item and relational memories.
problem Limited memory interactions in neural networks.
method Introduces a Self-attentive Associative Memory (SAM) operator to separate item and relational memories.
result Achieves competitive results in various tasks, including geometry, graph, reinforcement learning, and question answering.
Sparse codes improve optimal control tasks with correlated inputs.
problem Optimal control tasks with correlated feature inputs.
method Used a sparse code to represent natural images in an optimal control task solved with neuro-dynamic programming.
result An over-complete sparse code increases memory capacity and learning speed beyond a complete code.
EAST compresses deep ConvNets for tiny memory nodes.
problem Memory constraints in tiny devices for deep ConvNets.
method Encoding-Aware Sparse Training (EAST) with adaptive group pruning and LZ4 weight encoding.
result EAST achieves deep memory compression with lower sparsity and higher accuracy.
DAM with MRL improves relational reasoning in MANNs.
problem Limited performance of associative memory networks on complex relational reasoning tasks.
method Distributed Associative Memory architecture with Memory Refreshing Loss.
result Enhanced relation reasoning performance of MANNs on long temporal sequence data.
Generative diffusion models mimic biological memory networks, encoding associative dynamics in deep neural weights.
problem Understanding long-term memory mechanisms in neuroscience and AI.
method Interpreting generative diffusion models as energy-based models and comparing them to Hopfield networks.
result Generative diffusion models can encode associative dynamics of Hopfield networks in deep neural weights.
The paper studies scaling laws for associative memory mechanisms.
problem Understanding and optimizing learning and memorization processes.
method High-dimensional matrices of outer products of embeddings, relating to transformer models. Derived scaling laws with sample and parameter sizes. Extensive numerical experiments.
result Precise scaling laws and statistical efficiency of estimators.
Learning to remember long sequences remains a challenging task for recurrent neural networks. Register memory and attention mechanisms were both proposed to resolve the issue with either high computational cost to retain memory differentiability, or by discounting the RNN representation learning towards encoding shorte…
The sparse inverse covariance estimation problem is commonly solved using an ℓ1-regularized Gaussian maximum likelihood estimator known as "graphical lasso", but its computational cost becomes prohibitive for large data sets. A recent line of results showed--under mild assumptions--that the graphical lasso esti…
BiSHop tackles tabular data challenges with sparse Hopfield layers.
problem Non-rotationally invariant data structure and feature sparsity in tabular data.
method Sequential column-wise and row-wise processing through interconnected directional learning modules with generalized sparse modern Hopfield layers.
result BiSHop surpasses current SOTA methods with significantly less hyperparameter tuning.
This work improves reinforcement learning with sparse rewards by following diverse past trajectories.
problem Challenges in reinforcement learning with sparse rewards and myopic behavior.
method Proposes a trajectory-conditioned policy to learn from a memory buffer of diverse past trajectories.
result Significantly outperforms existing methods on complex tasks with local optima.
Comprehending complex systems by simplifying and highlighting important dynamical patterns requires modeling and mapping higher-order network flows. However, complex systems come in many forms and demand a range of representations, including memory and multilayer networks, which in turn call for versatile community-det…
SWAT reduces training efficiency and speed for CNNs.
problem Efficient training of neural networks on sparse hardware.
method SWAT modifies backpropagation to eliminate small weights and activations.
result SWAT reduces FLOPS by 80% on ResNet-50 for ImageNet, improving training speed by 3.3x.
Hydra boosts efficiency for long-context reasoning in resource-constrained settings.
problem Quadratic complexity of transformers limits long-context reasoning in resource-constrained systems.
method Hydra uses a modular architecture with adaptive routing between sparse global attention, mixture-of-experts, and dual memories.
result Hydra achieves significant throughput and accuracy improvements for long-context reasoning.
Sparse Meta Networks adapt deep neural networks incrementally for fast learning.
problem Training deep neural networks is slow and impractical for complex, changing environments.
method Sparse Meta Networks use a memory layer to learn online sequential adaptation, accumulating fast-weights incrementally.
result Sparse Meta Networks achieve strong performance in various sequential adaptation scenarios.
Recurrent Neural Networks (RNNs) are used in state-of-the-art models in domains such as speech recognition, machine translation, and language modelling. Sparsity is a technique to reduce compute and memory requirements of deep learning models. Sparse RNNs are easier to deploy on devices and high-end server processors. …
LLMs can be tricked into recalling facts based on context clues.
problem Manipulation of LLMs' factual recall through context changes.
method Mathematical exploration of transformers' associative memory properties.
result Transformers use self-attention and value matrix for associative memory.
This paper introduces a hierarchical associative memory model with multiple layers.
problem Limitations of traditional associative memory models with only one hidden layer.
method Develops a fully recurrent model with arbitrary layers, including locally connected ones, and a corresponding energy function.
result The model can dynamically assemble memories using weights from lower layers and higher layers' rules.
Meta-learning approach for fast and compressive energy-based memory models.
problem Learning associative memory with fast and compressive models for complex data.
method Meta-learning approach to energy-based memory models (EBMM) using arbitrary neural architectures.
result Demonstrated associative retrieval outperforming existing systems in reconstruction error and compression rate.
Sigmoid autoencoders can implement associative memory with certain conditions.
problem Implementing associative memory in neural networks.
method Theoretical analysis of overparameterized sigmoid autoencoders using the NTK and iterative maps.
result Overparameterized sigmoid autoencoders can have attractors in the NTK limit, leading to associative memory.
DMs emerge from DenseAMs, transitioning from memorization to generalization.
problem Hindered memory retrieval in DenseAMs due to spurious states.
method Examined diffusion models through the lens of DenseAMs, focusing on their generative process.
result Identified a critical phase in DMs transitioning from memorization to generalization.
SGM combines deep learning and planning for robust long-horizon tasks.
problem Combining deep learning and planning for robust long-horizon tasks.
method Sparse Graphical Memory (SGM) that stores states and feasible transitions in a sparse memory, aggregating states according to a two-way consistency objective.
result SGM significantly outperforms current state of the art methods on long horizon, sparse-reward visual navigation tasks.
Efficient tensor kernel method reduces memory usage and computational cost for sparse regression.
problem Memory and computational limitations in tensor kernel methods for sparse regression.
method Proposes a new tensor data layout and Nystrom subsampling approach to reduce memory and computational requirements.
result Improvements lead to more efficient tensor kernel methods for sparse regression.
Reverse Experience Replay improves Deep Q-learning for sparse rewards.
problem Sparse rewards and reward-maximizing tasks in Deep Q-learning.
method Sampling transitions in reverse order for training.
result Significantly increased performance in tasks with limited experience and memory capacity.
Learning long-term dependencies in extended temporal sequences requires credit assignment to events far back in the past. The most common method for training recurrent neural networks, back-propagation through time (BPTT), requires credit information to be propagated backwards through every single step of the forward c…
Transformers can store facts efficiently using associative memories.
problem Understanding how transformers store and recall factual information.
method Proved linear scaling of storage capacities for linear and MLP associative memories, introduced a synthetic task, and analyzed gradient flow.
result Shallow transformers can achieve near optimal storage capacity for factual recall tasks using associative memories.
An associative memory is a framework of content-addressable memory that stores a collection of message vectors (or a dataset) over a neural network while enabling a neurally feasible mechanism to recover any message in the dataset from its noisy version. Designing an associative memory requires addressing two main task…
A new memory system handles non-stationary environments by self-sizing and retaining memories.
problem Non-stationary environments where memories arrive over time and must be distinguished from noise.
method A self-sizing continual associative memory that generalizes Hopfield's network, handling adaptation and novelty.
result The memory system grows to the intrinsic memory demand of the environment and retains memories without forgetting.
A new method reduces memory requirements for Graph Transformers by sparsely training a network.
problem Quadratic memory complexity in Graph Transformers limits their scalability to large graphs.
method Spexphormer: trains a narrow network on augmented graph, then uses only active connections in a wider network.
result Spexphormer achieves good performance with drastically reduced memory requirements.
A new method selects features efficiently for high-dimensional data.
problem High computational costs and memory requirements in high-dimensional data.
method QuickSelection uses the strength of neurons in sparse autoencoders to select features.
result QuickSelection achieves the best trade-off of accuracy, speed, and memory usage.
Big T-Rex solves FDR-controlled sparse regression on laptops with millions of variables.
problem Scalable FDR-controlled variable selection for high-dimensional data.
method Early terminated random experiments with memory-mapping and permutation-based dummy generation.
result Solves FDR-controlled Lasso problems with 5 million variables on a laptop in 30 minutes.
A model of associative memory is studied, which stores and reliably retrieves many more patterns than the number of neurons in the network. We propose a simple duality between this dense associative memory and neural networks commonly used in deep learning. On the associative memory side of this duality, a family of mo…
Proposes RBGP framework for efficient block sparse neural networks.
problem Efficiently exploit structured sparsity patterns for sparse neural networks on GPU.
method Uses Ramanujan Bipartite Graph Product to generate structured multi-level block sparse neural networks.
result Achieves 5-9x and 2-5x runtime gains over unstructured and block sparsity patterns respectively, while maintaining accuracy.