paper-with-me

홈 › Papers

Beyond Scaling Laws: Understanding Transformer Performance with Associative Memory

2024-05-14 · Xueyan Niu, Bo Bai, Lei Deng, Wei Han

Increasing the size of a Transformer does not always lead to enhanced performance. This phenomenon cannot be explained by the empirical scaling laws. Furthermore, the model's enhanced performance is closely associated with its memorization of the training samples. We present a theoretical framework that sheds light on the memorization during pre-training of transformer-based language models. We model the behavior of Transformers with associative memories using Hopfield networks, such that each transformer block effectively conducts an approximate nearest-neighbor search. In particular, the energy function in modern continuous Hopfield networks serves as an explanation for the attention mechanism, which we approximate with a distance-based energy function. By observing that the softmax function corresponds to the gradient of the LogSumExp function in the energy, and employing the majorization-minimization technique, we construct a global energy function designed to capture the layered architecture. We demonstrate a dependency between the model size and the dataset size for the model to achieve optimal performance, and we show that the achievable cross-entropy loss is bounded from below.

📄 PDF Abstract BibTeX arXiv:2405.08707

Code (0)

등록된 구현이 없습니다.

Tasks

Memorization

Methods 이 논문이 사용한 방법론

Refunds@Expedia|||How do I get a full refund from Expedia? “How do I get a full refund from Expedia? How do I get a full refund from Expedia? – Call ☎️ +1-(888) 829 (0881) or +1-805-330-4056 or +1-805-330-4056 for Quick Help &…
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Discriminative Fine-Tuning Discriminative Fine-Tuning is a fine-tuning strategy that is used for ULMFiT type models. Instead of using the same learning rate…
Multi-Head Attention 설명 없음
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…
Attention Dropout Attention Dropout is a type of dropout used in attention-based architectures, where elements are randomly dropped out of the…
Position-Wise Feed-Forward Layer 설명 없음
Weight Decay 설명 없음

Similar Papers 제목 키워드 기반

Understanding Scaling Laws with Statistical and Approximation Theory for Transformer Neural Networks on Intrinsically Low-dimensional Data

2024-11-11 · Alex Havrilla, Wenjing Liao

When training deep neural networks, a model's generalization error is often observed to follow a power scaling law dependent both on the model size and the data size. Perhaps the best known example of such scaling laws a…

Scaling Laws Under the Microscope: Predicting Transformer Performance from Small Scale Experiments

2022-02-13 · Maor Ivgi, Yair Carmon, Jonathan Berant

Neural scaling laws define a predictable relationship between a model's parameter count and its performance after training in the form of a power law. However, most research to date has not explicitly investigated whethe…

Model Selection

Towards Neural Scaling Laws for Time Series Foundation Models

2024-10-16 · Qingren Yao, Chao-Han Huck Yang, Renhe Jiang, Yuxuan Liang 외

Scaling laws offer valuable insights into the design of time series foundation models (TSFMs). However, previous research has largely focused on the scaling laws of TSFMs for in-distribution (ID) data, leaving their out-…

DecoderTime Series

Scaling Laws for Autoregressive Generative Modeling

2020-10-28 · Tom Henighan, Jared Kaplan, Mor Katz, Mark Chen 외

We identify empirical scaling laws for the cross-entropy loss in four domains: generative image modeling, video modeling, multimodal image$\leftrightarrow$text models, and mathematical problem solving. In all cases autor…

Mathematical Problem-Solving

Diversity of Transformer Layers: One Aspect of Parameter Scaling Laws

2025-05-29 · Hidetaka Kamigaito, Ying Zhang, Jingun Kwon, Katsuhiko Hayashi 외

Transformers deliver outstanding performance across a wide range of tasks and are now a dominant backbone architecture for large language models (LLMs). Their task-solving performance is improved by increasing parameter …

Diversity