Reducing Model Jitter: Stable Re-training of Semantic Parsers in
Production Environments
- URL: http://arxiv.org/abs/2204.04735v1
- Date: Sun, 10 Apr 2022 17:57:55 GMT
- Title: Reducing Model Jitter: Stable Re-training of Semantic Parsers in
Production Environments
- Authors: Christopher Hidey, Fei Liu, Rahul Goel
- Abstract summary: Retraining modern deep learning systems can lead to variations in model performance even when trained using the same data and hyper- parameters.
We demonstrate the effectiveness of various jitter reduction techniques such as ensembling and distillation.
We show that co-distillation provides a sweet spot in terms of jitter reduction for semantic parsing systems with only a modest increase in resource usage.
- Score: 14.829119556960066
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: Retraining modern deep learning systems can lead to variations in model
performance even when trained using the same data and hyper-parameters by
simply using different random seeds. We call this phenomenon model jitter. This
issue is often exacerbated in production settings, where models are retrained
on noisy data. In this work we tackle the problem of stable retraining with a
focus on conversational semantic parsers. We first quantify the model jitter
problem by introducing the model agreement metric and showing the variation
with dataset noise and model sizes. We then demonstrate the effectiveness of
various jitter reduction techniques such as ensembling and distillation.
Lastly, we discuss practical trade-offs between such techniques and show that
co-distillation provides a sweet spot in terms of jitter reduction for semantic
parsing systems with only a modest increase in resource usage.
Related papers
- Towards Theoretical Understandings of Self-Consuming Generative Models [56.84592466204185]
This paper tackles the emerging challenge of training generative models within a self-consuming loop.
We construct a theoretical framework to rigorously evaluate how this training procedure impacts the data distributions learned by future models.
We present results for kernel density estimation, delivering nuanced insights such as the impact of mixed data training on error propagation.
arXiv Detail & Related papers (2024-02-19T02:08:09Z) - One More Step: A Versatile Plug-and-Play Module for Rectifying Diffusion
Schedule Flaws and Enhancing Low-Frequency Controls [77.42510898755037]
One More Step (OMS) is a compact network that incorporates an additional simple yet effective step during inference.
OMS elevates image fidelity and harmonizes the dichotomy between training and inference, while preserving original model parameters.
Once trained, various pre-trained diffusion models with the same latent domain can share the same OMS module.
arXiv Detail & Related papers (2023-11-27T12:02:42Z) - ExposureDiffusion: Learning to Expose for Low-light Image Enhancement [87.08496758469835]
This work addresses the issue by seamlessly integrating a diffusion model with a physics-based exposure model.
Our method obtains significantly improved performance and reduced inference time compared with vanilla diffusion models.
The proposed framework can work with both real-paired datasets, SOTA noise models, and different backbone networks.
arXiv Detail & Related papers (2023-07-15T04:48:35Z) - BOOT: Data-free Distillation of Denoising Diffusion Models with
Bootstrapping [64.54271680071373]
Diffusion models have demonstrated excellent potential for generating diverse images.
Knowledge distillation has been recently proposed as a remedy that can reduce the number of inference steps to one or a few.
We present a novel technique called BOOT, that overcomes limitations with an efficient data-free distillation algorithm.
arXiv Detail & Related papers (2023-06-08T20:30:55Z) - Empowering Diffusion Models on the Embedding Space for Text Generation [38.664533078347304]
We study the optimization challenges encountered with both the embedding space and the denoising model.
Data distribution is learnable for embeddings, which may lead to the collapse of the embedding space and unstable training.
Based on the above analysis, we propose Difformer, an embedding diffusion model based on Transformer.
arXiv Detail & Related papers (2022-12-19T12:44:25Z) - Distributionally Robust Models with Parametric Likelihood Ratios [123.05074253513935]
Three simple ideas allow us to train models with DRO using a broader class of parametric likelihood ratios.
We find that models trained with the resulting parametric adversaries are consistently more robust to subpopulation shifts when compared to other DRO approaches.
arXiv Detail & Related papers (2022-04-13T12:43:12Z) - Hierarchical model reduction driven by machine learning for parametric
advection-diffusion-reaction problems in the presence of noisy data [0.0]
We propose a new approach to generate a reliable reduced model for a parametric elliptic problem in the presence of noisy data.
We show that directional HiPOD looses in terms of accuracy when problem data are affected by noise.
We replace with Machine Learning fitting models which better discriminate relevant physical features in the data from irrelevant noise.
arXiv Detail & Related papers (2022-04-01T16:02:05Z) - Robust Training under Label Noise by Over-parameterization [41.03008228953627]
We propose a principled approach for robust training of over-parameterized deep networks in classification tasks where a proportion of training labels are corrupted.
The main idea is yet very simple: label noise is sparse and incoherent with the network learned from clean data, so we model the noise and learn to separate it from the data.
Remarkably, when trained using such a simple method in practice, we demonstrate state-of-the-art test accuracy against label noise on a variety of real datasets.
arXiv Detail & Related papers (2022-02-28T18:50:10Z) - Churn Reduction via Distillation [54.5952282395487]
We show an equivalence between training with distillation using the base model as the teacher and training with an explicit constraint on the predictive churn.
We then show that distillation performs strongly for low churn training against a number of recent baselines.
arXiv Detail & Related papers (2021-06-04T18:03:31Z) - Generative Text Modeling through Short Run Inference [47.73892773331617]
The present work proposes a short run dynamics for inference. It is variation from the prior distribution of the latent variable and then runs a small number of Langevin dynamics steps guided by its posterior distribution.
We show that the models trained with short run dynamics more accurately model the data, compared to strong language model and VAE baselines, and exhibit no sign of posterior collapse.
arXiv Detail & Related papers (2021-05-27T09:14:35Z)
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.