paper-with-me

홈 › Papers

Attention that does not Explain Away

2020-09-29 · Nan Ding, Xinjie Fan, Zhenzhong Lan, Dale Schuurmans, Radu Soricut

Models based on the Transformer architecture have achieved better accuracy than the ones based on competing architectures for a large set of tasks. A unique feature of the Transformer is its universal application of a self-attention mechanism, which allows for free information flow at arbitrary distances. Following a probabilistic view of the attention via the Gaussian mixture model, we find empirical evidence that the Transformer attention tends to "explain away" certain input neurons. To compensate for this, we propose a doubly-normalized attention scheme that is simple to implement and provides theoretical guarantees for avoiding the "explaining away" effect without introducing significant computational or memory cost. Empirically, we show that the new attention schemes result in improved performance on several well-known benchmarks.

📄 PDF Abstract BibTeX arXiv:2009.14308

Code (0)

등록된 구현이 없습니다.

Methods 이 논문이 사용한 방법론

Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…
Position-Wise Feed-Forward Layer 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…
Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…
BPE Byte Pair Encoding, or BPE, is a subword segmentation algorithm that encodes rare and unknown words as sequences of subword units. The intuition is that various word…
Label Smoothing Label Smoothing is a regularization technique that introduces noise for the labels. This accounts for the fact that datasets may have mistakes in them, so maximizing the…

Similar Papers 제목 키워드 기반

A Tutorial on Principal Component Analysis

2014-04-03 · Jonathon Shlens

Principal component analysis (PCA) is a mainstay of modern data analysis - a black box that is widely used but (sometimes) poorly understood. The goal of this paper is to dispel the magic behind this black box. This manu…

Runaway signals: Exaggerated displays of commitment may result from second-order signaling

2023-01-09 · Julien Lie-Panis, Jean-Louis Dessalles

To demonstrate their commitment, for instance during wartime, members of a group will sometimes all engage in the same ruinous display. Such uniform, high-cost signals are hard to reconcile with standard models of signal…

Grokking as Structural Inference: Transformers Need Bayesian Lottery Tickets

2026-05-15 · Kai Hidajat, Solden Stoll, Joseph An arxiv

Why does a Transformer that has memorized its training set wait thousands of steps before it generalizes? Existing accounts locate this delay in norm minimization, feature emergence, or the late discovery of sparse subne…

When does Privileged Information Explain Away Label Noise?

2023-03-03 · Guillermo Ortiz-Jimenez, Mark Collier, Anant Nawalgaria, Alexander D'Amour 외

Leveraging privileged information (PI), or features available during training but not at test time, has recently been shown to be an effective method for addressing label noise. However, the reasons for its effectiveness…

Explaining Away Attacks Against Neural Networks

2020-03-06 · Sean Saito, Jin Wang

We investigate the problem of identifying adversarial attacks on image-based neural networks. We present intriguing experimental results showing significant discrepancies between the explanations generated for the predic…