New memory in neural networks mimics computer architectures.
problem Learning algorithms and complex tasks with neural networks.
method Introducing a new memory to store weights for a neural controller, similar to stored-program memory in computers.
result Neural Stored-program Memory enhances neural networks' adaptability and learning capabilities.
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.
Dual memory improves reinforcement learning efficiency.
problem Training inefficiency in reinforcement learning.
method Introduces a dual memory structure with a main and cache memory.
result Dual memory structure leads to higher scores in reinforcement learning environments.
Stable Hadamard Memory improves reinforcement learning by efficiently managing memory.
problem Memory models struggle in partially observable reinforcement learning environments.
method Introduces a novel memory model using the Hadamard product for efficient memory management and updates.
result Significantly outperforms state-of-the-art memory-based methods on challenging benchmarks.
ARMIN improves memory efficiency and lightness in neural networks.
problem Complex memory addressing and inefficient memory information exploitation in previous MANNs.
method ARMIN uses hidden state for automatic memory addressing and a novel RNN cell for memory integration.
result ARMIN achieves lower computational overhead and similar performances compared to vanilla LSTM.
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.
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.
We discuss memory models which are based on tensor decompositions using latent representations of entities and events. We show how episodic memory and semantic memory can be realized and discuss how new memory traces can be generated from sensory input: Existing memories are the basis for perception and new memories ar…
mGRN improves multivariate time series prediction by managing marginal and joint memories.
problem Extracting dependencies in multivariate sequential data with strong serial and cross-sectional dependencies.
method Developed a novel recurrent network architecture, Memory-Gated Recurrent Networks (mGRN), with gates for marginal and joint memories.
result mGRN consistently outperforms state-of-the-art architectures on various public datasets.
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.
Metalearned neural memory improves learning across various tasks.
problem Improving neural network adaptability and memory function.
method Augmenting neural networks with a metalearned external memory mechanism.
result The model achieves strong performance on diverse learning problems.
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…
Enhanced Hopfield model boosts memory retrieval capacity.
problem Memory retrieval in modern Hopfield models with limited capacity.
method Introduces a learnable feature map transforming energy function into kernel space, minimizing separation loss for uniform memory distribution.
result Significant reduction in metastable states, enhancing memory capacity and retrieval accuracy.
Current generation of memory-augmented neural networks has limited scalability as they cannot efficiently process data that are too large to fit in the external memory storage. One example of this is lifelong learning scenario where the model receives unlimited length of data stream as an input which contains vast majo…
Memory networks are neural networks with an explicit memory component that can be both read and written to by the network. The memory is often addressed in a soft way using a softmax function, making end-to-end training with backpropagation possible. However, this is not computationally scalable for applications which …
Memory-constrained algorithms need superlinear memory for efficient convex optimization.
problem Efficiently minimizing convex functions with limited memory.
method Analyzing first-order algorithms with superlinear memory constraints.
result Superlinear memory is necessary for optimal performance in convex optimization.
We present an end-to-end trained memory system that quickly adapts to new data and generates samples like them. Inspired by Kanerva's sparse distributed memory, it has a robust distributed reading and writing mechanism. The memory is analytically tractable, which enables optimal on-line compression via a Bayesian updat…
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.
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.
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.
Accelerators with power-law memory are proposed in the framework of the discrete time approach. To describe discrete accelerators we use the capital stock adjustment principle, which has been suggested by Matthews.The suggested discrete accelerators with memory describe the economic processes with the power-law memory …
Paper questions RNN and LSTM's long-term memory and introduces a new definition.
problem Whether RNN and LSTM have long-term memory.
method Introduced a new definition of long-term memory and modified RNN and LSTM to test it.
result RNN and LSTM do not meet the new definition of long-term memory.
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…
Augmenting a neural network with memory that can grow without growing the number of trained parameters is a recent powerful concept with many exciting applications. We propose a design of memory augmented neural networks (MANNs) called Labeled Memory Networks (LMNs) suited for tasks requiring online adaptation in class…
We study the long-term memory in diverse stock market indices and foreign exchange rates using the Detrended Fluctuation Analysis(DFA). For all daily and high-frequency market data studied, no significant long-term memory property is detected in the return series, while a strong long-term memory property is found in th…
New memory effect discovered in gravitational wave behavior.
problem Understanding gravitational wave behavior in spacetimes with angular momentum.
method Mathematical analysis of Minkowski spacetime and Kerr black holes.
result Angular momentum memory effect observed at future null infinity.
MEMGAN uses memory to improve anomaly detection by isolating abnormal data.
problem Weak guarantees for detecting anomalous data in classical algorithms.
method Memory-augmented Generative Adversarial Networks (MEMGAN) with a memory module.
result MEMGAN provides strong guarantees for anomaly detection with improved reconstruction.
Memory is increasingly often the bottleneck when training neural network models. Despite this, techniques to lower the overall memory requirements of training have been less widely studied compared to the extensive literature on reducing the memory requirements of inference. In this paper we study a fundamental questio…
Tensor models decode human perception and memory using SPO triples.
problem Understanding implicit and explicit perception and memory in the brain.
method Tensor models with SPO triples, dual representations, and four layers.
result Semantic memory is crucial for explicit perception and declarative memories.
Paper proposes low-rank gradient approximation to save memory for deep neural network training.
problem Memory limitation on mobile devices for deep neural network training.
method Approximating gradient matrices using low-rank parameterization.
result Reduces training memory by about 33.0% for Adam optimization and 4.5% relative lower word error rate on ASR personalization task.
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.
We consider a novel question answering (QA) task where the machine needs to read from large streaming data (long documents or videos) without knowing when the questions will be given, which is difficult to solve with existing QA methods due to their lack of scalability. To tackle this problem, we propose a novel end-to…
Our research is focused on understanding and applying biological memory transfers to new AI systems that can fundamentally improve their performance, throughout their fielded lifetime experience. We leverage current understanding of biological memory transfer to arrive at AI algorithms for memory consolidation and repl…
Study asks if memory constraints affect optimal convex optimization methods.
problem Characterize the minimax number of queries for convex optimization with memory constraints.
method Analyze first order methods under memory limitations.
result Optimal oracle complexity may be achievable with limited memory.
Active-memory mechanisms can replace self-attention in Transformers, but optimal results often require both.
problem Replacing self-attention with active-memory mechanisms in Transformers.
method Evaluation of various active-memory mechanisms in a Transformer model.
result Active-memory mechanisms can achieve comparable results to self-attention for language modeling, but optimal results are often achieved by combining both mechanisms.
A novel memory mechanism for reinforcement learning agents that stores past events in human-readable language.
problem Lack of interpretability in reinforcement learning agent's memory mechanisms.
method Uses CLIP to associate visual inputs with language tokens, then feeds these tokens to a pretrained language model.
result Significantly faster convergence on challenging continuous recognition tasks.
Unified model explains volatility memory in stocks and forex.
problem Understanding the components of volatility memory in financial markets.
method Developed a three-dimensional decomposition of volatility memory into level, shape, and tempo.
result Unified model shows that volatility memory is state-dependent, with different gates prevailing in equities and forex.
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.
Enhanced image recognition models learn from human-like memory and shape biases.
problem Improving robustness of image recognition models against various perturbations.
method Integrating human-like episodic memory and shape bias features into image recognition models.
result Combining human-like features improves robustness against both adversarial and natural perturbations.
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.
We design and study a Contextual Memory Tree (CMT), a learning memory controller that inserts new memories into an experience store of unbounded size. It is designed to efficiently query for memories from that store, supporting logarithmic time insertion and retrieval operations. Hence CMT can be integrated into existi…
A technique identifies memoryless algorithms approximating memory-dependent optimization methods.
problem Understanding how memory in optimization algorithms affects loss and generalization.
method Introducing a general technique to replace past iterates with the current one and adding a correction term.
result Lion does not have the same implicit anti-regularization as AdamW, explaining its better generalization performance.
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…
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.
FRD protects privacy in distributed RL by sharing proxy experience memory.
problem Privacy violation in exchanging experience memory in distributed RL.
method Proposes FRD framework using proxy experience memory.
result Numerical evaluation shows FRD is effective and performance depends on proxy memory structure.
New analysis tightens memory capacity of Hopfield models using spherical codes.
problem Optimizing memory capacity in modern Hopfield models and Kernelized Hopfield Models.
method Connecting Hopfield models to spherical codes in information theory, establishing an optimal capacity bound and a sub-linear algorithm.
result First tight and optimal asymptotic memory capacity for modern Hopfield models, matching known lower bounds.
MemNet optimizes neural architectures for memory efficiency.
problem Memory constraints in mobile devices limit the use of large neural networks.
method Augment-trim learning with memory consumption ranking score.
result MemNet finds architectures with 24.17% less memory usage compared to state-of-the-art methods.
New memory models improve LSSVM's generalization and reduce time cost.
problem Improving LSSVM's generalization performance and reducing training time.
method Introducing MIMM and WIMM models that modify LSSVM's memory influence mechanism.
result MIMM and WIMM models outperform LSSVM in generalization and time efficiency.