Transformers learn in-context by gradient descent

• URL: http://arxiv.org/abs/2212.07677v2
• Date: Wed, 31 May 2023 08:59:47 GMT
• Title: Transformers learn in-context by gradient descent
• Authors: Johannes von Oswald, Eyvind Niklasson, Ettore Randazzo, Jo\~ao Sacramento, Alexander Mordvintsev, Andrey Zhmoginov, Max Vladymyrov
• Abstract summary: Training Transformers on auto-regressive objectives is closely related to gradient-based meta-learning formulations. We show how trained Transformers become mesa-optimizers i.e. learn models by gradient descent in their forward pass.
• Score: 58.24152335931036
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: At present, the mechanisms of in-context learning in Transformers are not well understood and remain mostly an intuition. In this paper, we suggest that training Transformers on auto-regressive objectives is closely related to gradient-based meta-learning formulations. We start by providing a simple weight construction that shows the equivalence of data transformations induced by 1) a single linear self-attention layer and by 2) gradient-descent (GD) on a regression loss. Motivated by that construction, we show empirically that when training self-attention-only Transformers on simple regression tasks either the models learned by GD and Transformers show great similarity or, remarkably, the weights found by optimization match the construction. Thus we show how trained Transformers become mesa-optimizers i.e. learn models by gradient descent in their forward pass. This allows us, at least in the domain of regression problems, to mechanistically understand the inner workings of in-context learning in optimized Transformers. Building on this insight, we furthermore identify how Transformers surpass the performance of plain gradient descent by learning an iterative curvature correction and learn linear models on deep data representations to solve non-linear regression tasks. Finally, we discuss intriguing parallels to a mechanism identified to be crucial for in-context learning termed induction-head (Olsson et al., 2022) and show how it could be understood as a specific case of in-context learning by gradient descent learning within Transformers. Code to reproduce the experiments can be found at https://github.com/google-research/self-organising-systems/tree/master/transformers_learn_icl_by_gd .

Related papers

• Can Looped Transformers Learn to Implement Multi-step Gradient Descent for In-context Learning? [69.4145579827826]
  We show a fast flow on the regression loss despite the gradient non-ity algorithms for our convergence landscape. This is the first theoretical analysis for multi-layer Transformer in this setting.
  arXiv  Detail & Related papers  (2024-10-10T18:29:05Z)
• Learning on Transformers is Provable Low-Rank and Sparse: A One-layer Analysis [63.66763657191476]
  We show that efficient numerical training and inference algorithms as low-rank computation have impressive performance for learning Transformer-based adaption. We analyze how magnitude-based models affect generalization while improving adaption. We conclude that proper magnitude-based has a slight on the testing performance.
  arXiv  Detail & Related papers  (2024-06-24T23:00:58Z)
• How Well Can Transformers Emulate In-context Newton's Method? [46.08521978754298]
  We study whether Transformers can perform higher order optimization methods, beyond the case of linear regression. We demonstrate the ability of even linear attention-only Transformers in implementing a single step of Newton's iteration for matrix inversion with merely two layers.
  arXiv  Detail & Related papers  (2024-03-05T18:20:10Z)
• Linear Transformers are Versatile In-Context Learners [19.988368693379087]
  We prove that each layer of a linear transformer maintains a weight vector for an implicit linear regression problem. We also investigate the use of linear transformers in a challenging scenario where the training data is corrupted with different levels of noise. Remarkably, we demonstrate that for this problem linear transformers discover an intricate and highly effective optimization algorithm.
  arXiv  Detail & Related papers  (2024-02-21T23:45:57Z)
• How do Transformers perform In-Context Autoregressive Learning? [76.18489638049545]
  We train a Transformer model on a simple next token prediction task. We show how a trained Transformer predicts the next token by first learning $W$ in-context, then applying a prediction mapping.
  arXiv  Detail & Related papers  (2024-02-08T16:24:44Z)
• Linear attention is (maybe) all you need (to understand transformer optimization) [55.81555204646486]
  We make progress towards understanding the subtleties of training Transformers by studying a simple yet canonicalized shallow Transformer model. Most importantly, we observe that our proposed linearized models can reproduce several prominent aspects of Transformer training dynamics.
  arXiv  Detail & Related papers  (2023-10-02T10:48:42Z)
• Transformers learn to implement preconditioned gradient descent for in-context learning [41.74394657009037]
  Several recent works demonstrate that transformers can implement algorithms like gradient descent. We ask: Can transformers learn to implement such algorithms by training over random problem instances? For a transformer with $L$ attention layers, we prove certain critical points of the training objective implement $L$ iterations of preconditioned gradient descent.
  arXiv  Detail & Related papers  (2023-06-01T02:35:57Z)
• Emergent Agentic Transformer from Chain of Hindsight Experience [96.56164427726203]
  We show that a simple transformer-based model performs competitively with both temporal-difference and imitation-learning-based approaches. This is the first time that a simple transformer-based model performs competitively with both temporal-difference and imitation-learning-based approaches.
  arXiv  Detail & Related papers  (2023-05-26T00:43:02Z)
• The Closeness of In-Context Learning and Weight Shifting for Softmax Regression [42.95984289657388]
  We study the in-context learning based on a softmax regression formulation. We show that when training self-attention-only Transformers for fundamental regression tasks, the models learned by gradient-descent and Transformers show great similarity.
  arXiv  Detail & Related papers  (2023-04-26T04:33:41Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.