paper-with-me

Papers

Memory-efficient Stochastic methods for Memory-based Transformers

2023-11-14 · Vishwajit Kumar Vishnu, C. Chandra Sekhar

Training Memory-based transformers can require a large amount of memory and can be quite inefficient. We propose a novel two-phase training mechanism and a novel regularization technique to improve the training efficiency of memory-based transformers, which are often used for long-range context problems. For our experiments, we consider transformer-XL as our baseline model which is one of memorybased transformer models. We show that our resultant model, Skip Cross-head TransformerXL, outperforms the baseline on character level language modeling task with similar parameters and outperforms the baseline on word level language modelling task with almost 20% fewer parameters. Our proposed methods do not require any additional memory. We also demonstrate the effectiveness of our regularization mechanism on BERT which shows similar performance with reduction in standard deviation of scores of around 30% on multiple GLUE tasks.

📄 PDF Abstract BibTeX arXiv:2311.08123

Code (1)

vishwajit-vishnu/memory-efficient-stochastic-methods-for-memory-based-transformers 공식 구현 pytorch

Tasks

Language ModelingLanguage ModellingParaphrase Identification

Methods 이 논문이 사용한 방법론

ReLU How Do I Communicate to Expedia? How Do I Communicate to Expedia? – Call ☎️ +1-(888) 829 (0881) or +1-805-330-4056 or +1-805-330-4056 for Live Support & Special Travel…
Multi-Head Attention 설명 없음
Attention 설명 없음
Cosine Annealing Cosine Annealing is a type of learning rate schedule that has the effect of starting with a large learning rate that is relatively rapidly decreased to a minimum value before…
Adaptive Softmax Adaptive Softmax is a speedup technique for the computation of probability distributions over words. The adaptive softmax is…
Adaptive Input Representations Adaptive Input Embeddings extend the adaptive softmax to input word representations. The factorization assigns more…
Linear Warmup With Cosine Annealing Linear Warmup With Cosine Annealing is a learning rate schedule where we increase the learning rate linearly for $n$ updates and then anneal according to a cosine schedule…
Variational Dropout Variational Dropout is a regularization technique based on dropout, but uses a variational inference grounded approach. In…

Similar Papers 제목 키워드 기반

ARTEMIS: A Mixed Analog-Stochastic In-DRAM Accelerator for Transformer Neural Networks

2024-07-17 · Salma Afifi, Ishan Thakkar, Sudeep Pasricha

Transformers have emerged as a powerful tool for natural language processing (NLP) and computer vision. Through the attention mechanism, these models have exhibited remarkable performance gains when compared to conventio…

CPUGPU

Winner-Take-All Column Row Sampling for Memory Efficient Adaptation of Language Model

2023-05-24 · NeurIPS 2023 11 · Zirui Liu, Guanchu Wang, Shaochen Zhong, Zhaozhuo Xu 외

With the rapid growth in model size, fine-tuning the large pre-trained language model has become increasingly difficult due to its extensive memory usage. Previous works usually focus on reducing the number of trainable …

AllLanguage ModelingLanguage ModellingStochastic Optimization

B'MOJO: Hybrid State Space Realizations of Foundation Models with Eidetic and Fading Memory

2024-07-08 · Luca Zancato, Arjun Seshadri, Yonatan Dukler, Aditya Golatkar 외

We describe a family of architectures to support transductive inference by allowing memory to grow to a finite but a-priori unknown bound while making efficient use of finite resources for inference. Current architecture…

Language ModelingLanguage ModellingMambaState Space Models

Sustainable Transformer Neural Network Acceleration with Stochastic Photonic Computing

2026-04-10 · S. Afifi, O. Alo, I. Thakkar, S. Pasricha arxiv

Transformers achieve state-of-the-art performance in natural language processing, vision, and scientific computing, but demand high computation and memory. To address these challenges, we present ASTRA, the first silicon…

No More Adam: Learning Rate Scaling at Initialization is All You Need

2024-12-16 · Minghao Xu, Lichuan Xiang, Xu Cai, Hongkai Wen

In this work, we question the necessity of adaptive gradient methods for training deep neural networks. SGD-SaI is a simple yet effective enhancement to stochastic gradient descent with momentum (SGDM). SGD-SaI performs …

All