Typical neural networks with external memory do not effectively separate capacity for episodic and working memory as is required for reasoning in humans. Applying knowledge gained from psychological studies, we designed a new model called Differentiable Working Memory (DWM) in order to specifically emulate human workin…
On-device research index
arXiv research
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.
168,695 papers · 148 categories
Trend · papers per month
48 results for “working memory”
Model combines long-term and short-term memory using conceptors.
problem Transfer between long-term and short-term memory.
method Recurrent neural network with gated reservoir for short-term memory and conceptors for long-term memory.
result Standard operations on conceptors allow combining long-term memories and describing their effect on short-term memory.
A new model for sequential memory using temporal predictive coding.
problem Forming accurate memory of sequential stimuli in the brain.
method Proposes a novel PC-based model called temporal predictive coding (tPC).
result Shows that tPC models can accurately memorize and retrieve sequential inputs.
The key to a Transformer model is the self-attention mechanism, which allows the model to analyze an entire sequence in a computationally efficient manner. Recent work has suggested the possibility that general attention mechanisms used by RNNs could be replaced by active-memory mechanisms. In this work, we evaluate wh…
This work analyzes and optimizes memory and compute costs of learned optimizers.
problem High memory and compute costs of learned optimizers.
method Identified and quantified design features of learned and hand-designed optimizers, constructed a more efficient learned optimizer.
result A learned optimizer that is faster and more memory efficient than previous work.
ADHD is being recognized as a diagnosis which persists into adulthood impacting economic, occupational, and educational outcomes. There is an increased need to accurately diagnose and recommend interventions for this population. One consideration is the development and implementation of reliable and valid outcome measu…
Low-rank training improves neural network training on edge devices with non-volatile memory.
problem Training neural networks on edge devices with non-volatile memory, especially in terms of write density and auxiliary memory.
method Low-rank training scheme to address write density and auxiliary memory limitations.
result The low-rank training technique outperforms standard SGD in accuracy and weight writes.
This work tackles online memory selection in continual learning using information theory.
problem Online selection of a representative replay memory from data streams.
method Information-theoretic criteria (surprise, learnability) and Bayesian model for efficient computation.
result InfoRS improves robustness against data imbalance compared to reservoir sampling.
Memory-limited learning tackles adversarial bandits with reduced storage.
problem Adversarial bandit problem with limited memory storage.
method Hierarchical learning policy with sublinear memory requirement.
result Established sublinear regret bounds for weak and shifting regrets.
This study examines memory effects in S&P500 market correlations using Langevin models.
problem The neglect of memory effects in market correlations for optimal portfolio selection.
method Fit a generalised Langevin equation (GLE) to S&P500 market correlation data.
result Memory effects in market correlations significantly improve forecasting accuracy and suggest a hidden slow time scale.
We propose a new architecture called Memory-Augmented Encoder-Solver (MAES) that enables transfer learning to solve complex working memory tasks adapted from cognitive psychology. It uses dual recurrent neural network controllers, inside the encoder and solver, respectively, that interface with a shared memory module a…
The ability to look multiple times through a series of pose-adjusted glimpses is fundamental to human vision. This critical faculty allows us to understand highly complex visual scenes. Short term memory plays an integral role in aggregating the information obtained from these glimpses and informing our interpretation …
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…
Kanerva++: extending The Kanerva Machine with differentiable, locally block allocated latent memorycs.NE
New memory allocation scheme improves image generation performance.
problem Improving episodic and semantic memory representation in neural networks.
method Developed a hierarchical latent variable model with differentiable, locally block allocated latent memory.
result Improved conditional likelihood values on various datasets.
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.
When analyzing the genome, researchers have discovered that proteins bind to DNA based on certain patterns of the DNA sequence known as "motifs". However, it is difficult to manually construct motifs due to their complexity. Recently, externally learned memory models have proven to be effective methods for reasoning ov…
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.
The implementation of Deep Convolutional Neural Networks (ConvNets) on tiny end-nodes with limited non-volatile memory space calls for smart compression strategies capable of shrinking the footprint yet preserving predictive accuracy. There exist a number of strategies for this purpose, from those that play with the to…
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.
In distributed reinforcement learning, it is common to exchange the experience memory of each agent and thereby collectively train their local models. The experience memory, however, contains all the preceding state observations and their corresponding policies of the host agent, which may violate the privacy of the ag…
NeuZip compresses neural network weights to save memory during training and inference.
problem Memory constraints in neural network training and inference.
method Entropy-based dynamic weight compression.
result Significant reduction in memory usage without performance loss.
Many robotic applications require the agent to perform long-horizon tasks in partially observable environments. In such applications, decision making at any step can depend on observations received far in the past. Hence, being able to properly memorize and utilize the long-term history is crucial. In this work, we pro…
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.
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.
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.
In this work we propose a new class of long-memory models with time-varying fractional parameter. In particular, the dynamics of the long-memory coefficient, , is specified through a stochastic recurrence equation driven by the score of the predictive likelihood, as suggested by Creal et al. (2013) and Harvey (2013)…
Memory is an important aspect of intelligence and plays a role in many deep reinforcement learning models. However, little progress has been made in understanding when specific memory systems help more than others and how well they generalize. The field also has yet to see a prevalent consistent and rigorous approach f…
New solver avoids memory issues for long differential equations.
problem Memory constraints in adaptive probabilistic ODE solvers.
method Fixed memory demands adaptive probabilistic solver using robust state estimation.
result Eliminates memory issues for long time series simulations.
Study reduces memory needs for active learning with enriched queries.
problem Expensive labeling costs in active learning.
method Introduces bounded memory active learning through enriched queries, introduces lossless sample compression.
result Can learn classifiers with bounded memory and query optimality.
We investigate the robustness properties of image recognition models equipped with two features inspired by human vision, an explicit episodic memory and a shape bias, at the ImageNet scale. As reported in previous work, we show that an explicit episodic memory improves the robustness of image recognition models agains…
A central challenge faced by memory systems is the robust retrieval of a stored pattern in the presence of interference due to other stored patterns and noise. A theoretically well-founded solution to robust retrieval is given by attractor dynamics, which iteratively clean up patterns during recall. However, incorporat…
NEST optimizes deep learning training by placing devices efficiently across networks and memory.
problem Inefficient device placement in distributed deep learning leads to high communication and memory overhead.
method NEST uses network-, compute-, and memory-aware dynamic programming to optimize device placement.
result NEST achieves up to 2.43 times higher throughput and better memory efficiency.
New neural operators model turbulence with memory and randomness.
problem Modeling turbulence in complex fluid dynamics with memory and randomness.
method Symmetrized activation functions, fractional derivatives, and stochastic noise.
result Theoretical guarantees for approximation quality in turbulent phenomena.
PipeDream-2BW accelerates large model training by 20x with minimal memory usage.
problem Training large models requires memory beyond single accelerator capacity.
method Pipeline parallelism, weight gradient coalescing, double buffering.
result Accelerates large model training by up to 20x.
This research unifies concepts of fading memory in RNNs.
problem Unclear relationships between fading memory concepts in RNNs.
method Unified language and new proofs for fading memory concepts.
result Clarified relationships between fading memory concepts.
New approach reduces malware detection memory requirements and speeds up training.
problem Efficiently classifying long sequences of malware detection data.
method Developed a new temporal max pooling method and global channel gating design.
result 116x more memory efficient and 25.8x faster training on original dataset.
Research on long-range memory in financial and social systems using various models.
problem Understanding the nature of long-range memory in socioeconomic systems.
method Various Markov processes including point processes, stochastic differential equations, and agent-based models.
result New estimators of self-similarity and long-range memory for non-Gaussian systems are needed.
SAGE improves memory efficiency by selectively adding, merging, or ignoring new facts.
problem Efficiently managing new facts in agentic LLMs to avoid costly write-time reasoning.
method SAGE uses a von Mises-Fisher-based density estimator to score and route candidate facts.
result SAGE achieves the best average token-F1 on LoCoMo and reduces add-phase API cost by 3.4x on GPT-4o-mini.
Enhances sequence memory capacity in neural networks.
problem Limited sequence capacity in Hopfield-like neural networks.
method Introducing a nonlinear interaction term and a generalized pseudoinverse rule.
result Significantly increased sequence capacity with novel scaling laws.
Memory-efficient learning for large-scale imaging systems.
problem Memory limitations in GPUs for real-world large-scale inverse problems.
method Exploits reversibility of network layers to enable data-driven design.
result Demonstrated on small-scale and large-scale real-world systems.
Deep learning typically requires training a very capable architecture using large datasets. However, many important learning problems demand an ability to draw valid inferences from small size datasets, and such problems pose a particular challenge for deep learning. In this regard, various researches on "meta-learning…
We describe a procedure which verifies that a group given by generators and relators is word-hyperbolic. This procedure always works with a group which is word-hyperbolic, provided there is sufficient memory and time devoted to the problem. If the group is not word-hyperbolic, the procedure continues indefinitely. We a…
Efficient echo state network with explicit memory performs well on benchmark tasks.
problem Training differentiable neural computers is difficult and time-consuming.
method Echo state network with an explicit memory.
result Echo state network can recognize all regular languages, including those contractive networks cannot.
Investment decisions can benefit from incorporating an accumulated knowledge of the past to drive future decision making. We introduce Continual Learning Augmentation (CLA) which is based on an explicit memory structure and a feed forward neural network (FFNN) base model and used to drive long term financial investment…
In continual learning (CL), an agent learns from a stream of tasks leveraging prior experience to transfer knowledge to future tasks. It is an ideal framework to decrease the amount of supervision in the existing learning algorithms. But for a successful knowledge transfer, the learner needs to remember how to perform …
Communication-Efficient, 2D Parallel Stochastic Gradient Descent for Distributed-Memory Optimizationcs.DC
HybridSGD improves SGD performance by balancing computation and communication.
problem Limited scalability and performance of SGD due to communication costs.
method 2D parallel SGD method (HybridSGD) that trades off between 1D -step SGD and 1D Federated SGD (FedAvg).
result HybridSGD achieves better convergence than FedAvg at similar processor scales and up to 121x speedup over FedAvg.
This work investigates power laws in deep neural network ensembles and predicts their performance.
problem Understanding the performance of deep neural network ensembles and their optimal structure.
method Investigated the behavior of negative log-likelihood (CNLL) of a deep ensemble as a function of ensemble size and member network size, identifying power law dependencies.
result One large network may perform worse than an ensemble of several medium-size networks, known as a memory split.
Sketchy reduces memory and compute requirements for adaptive regularization in deep learning.
problem Prohibitive memory and running time for adaptive regularization methods in deep learning.
method Low-rank sketching approach using Frequent Directions (FD) to reduce memory and compute requirements.
result Efficient interpolation between resource requirements and degradation in regret guarantees with rank .