Françoise Beaufays
Françoise Beaufays is a Research Scientist at Google, where she leads a team working on on-device machine learning for Speech and Mobile Keyboard models. Her area of scientific expertise covers deep learning, sequence-to-sequence modeling, language modeling and other technologies related to natural language processing, with a recent focus on privacy-preserving modeling techniques. Françoise studied Mechanical and Electrical Engineering in Brussels, Belgium. She holds a PhD in Electrical Engineering and a PhD minor in Italian Literature, both from Stanford University.
Authored Publications
Sort By
Online Model Compression for Federated Learning with Large Models
Tien-Ju Yang
Giovanni Motta
(2022)
Preview abstract
This paper addresses the challenges of training large neural network models under federated learning settings: high on-device memory usage and communication cost. The proposed Online Model Compression (OMC) provides a framework that stores model parameters in a compressed format and decompresses them only when needed. We use quantization as the compression method in this paper and propose three methods, (1) using per-variable transformation, (2) weight matrices only quantization, and (3) partial parameter quantization, to minimize the impact on model accuracy. According to our experiments on two recent neural networks for speech recognition and two different datasets, OMC can reduce memory usage and communication cost of model parameters by up to 59% while attaining comparable accuracy and training speed when compared with full-precision training.
View details
A Method to Reveal Speaker Identity in Distributed ASR Training,and How to Counter It
Trung Dang
Peter Chin
IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2022, Virtual and Singapore, 23-27 May 2022, {IEEE}, pp. 4338-4342
Preview abstract
End-to-end Automatic Speech Recognition (ASR) models are commonly trained over spoken utterances using optimization methods like Stochastic Gradient Descent (SGD). In distributed settings like Federated Learning, model training requires transmission of gradients over a network. In this work, we design the first method for revealing the identity of the speaker of a training utterance with access only to a gradient. We propose Hessian-Free Gradients Matching, an input reconstruction technique that operates without second derivatives of the loss function (required in prior works), which can be expensive to compute. We show the effectiveness of our method using the DeepSpeech model architecture, demonstrating that it is possible to reveal the speaker’s identity with 34% top-1 accuracy (51% top-5 accuracy) on the LibriSpeech dataset. Further, we study the effect of Dropout on the success of our method. We show that a dropout rate of 0.2 can reduce the speaker identity accuracy to 0% top-1 (0.5% top-5).
View details
Federated Pruning: Improving Neural Network Efficiency with Federated Learning
Rongmei Lin
Tien-Ju Yang
Ding Zhao
Li Xiong
Giovanni Motta
Interspeech 2022(2022)
Preview abstract
Automatic Speech Recognition models require large amount of speech data for training, and the collection of such data often leads to privacy concerns. Federated learning has been widely used and is considered to be an effective decentralized technique by collaboratively learning a shared prediction model while keeping the data local on different clients devices. However, the limited computation and communication resources on clients devices present practical difficulties for large models. To overcome such challenges, we propose Federated Pruning to train a reduced model under the federated setting, while maintaining similar performance compared to the full model. Moreover, the vast amount of clients data can also be leveraged to improve the pruning results compared to centralized training. We explore different pruning schemes and provide empirical evidence of the effectiveness of our methods.
View details
Handling Compounding in Mobile Keyboard Input
Andreas Christian Kabel
Keith B. Hall
David Rybach
arXiv cs.CL(2022)
Preview abstract
This paper proposes a framework to improve the typing experience of mobile users in morphologically rich languages. Smartphone keyboards typically support features such as input decoding, corrections and predictions that all rely on language models. For latency reasons, these operations happen on device, so the models are of limited size and cannot easily cover all the words needed by users for their daily tasks, especially in morphologically rich languages. In particular, the compounding nature of Germanic languages makes their vocabulary virtually infinite. Similarly, heavily inflecting and agglutinative languages (e.g. Slavic, Turkic or Finno-Ugric languages) tend to have much larger vocabularies than morphologically simpler languages, such as English or Mandarin. We propose to model such languages with automatically selected subword units annotated with what we call binding types, allowing the decoder to know when to bind subword units into words. We show that this method brings around 20% word error rate reduction in a variety of compounding languages. This is more than twice the improvement we previously obtained with a more basic approach, also described in the paper.
View details
Large-scale ASR Domain Adaptation by Self- and Semi-supervised Learning
David Qiu
Nikhil Siddhartha
ICASSP(2022) (to appear)
Preview abstract
Self- and Semi-supervised learning methods have been actively investigated to reduce labeled training data or enhance the model performance. However, the approach mostly focus on in-domain performance for public datasets. In this study, we utilize the combination of self- and semi-supervised learning methods to solve unseen domain adaptation problem in a large-scale production setting for online ASR model. This approach demonstrates that using the source domain data with a small fraction of the target domain data (3%) can recover the performance gap compared to a full data baseline: relative 13.5% WER improvement for target domain data.
View details
Extracting Targeted Training Data from ASR Models, and How to Mitigate It
Ehsan Amid
Proc. Interspeech 2022(2022) (to appear)
Preview abstract
Recent work has designed methods to demonstrate that model updates in ASR training can leak potentially sensitive attributes of the utterances used in computing the updates. In this work, we design the first method to demonstrate information leakage about training data from trained ASR models. We design Noise Masking, a fill-in-the-blank style method for extracting targeted parts of training data from trained ASR models. We demonstrate the success of Noise Masking by using it in four settings for extracting names from the LibriSpeech dataset used for training a state-of-the-art Conformer model. In particular, we show that we are able to extract the correct names from masked training utterances with 11.8% accuracy, while the model outputs some name from the train set 55.2% of the time. Further, we show that even in a setting that uses synthetic audio and partial transcripts from the test set, our method achieves 2.5% correct name accuracy (47.7% any name success rate). Lastly, we design Word Dropout, a data augmentation method that we show when used in training along with Multistyle TRaining (MTR), provides comparable utility as the baseline, along with significantly mitigating extraction via Noise Masking across the four evaluated settings.
View details
Revealing and Protecting Labels in Distributed Training
Trung Dang
Peter Chin
Advances in Neural Information Processing Systems 34: Annual Conference on Neural Information Processing Systems 2021, NeurIPS 2021, December 6-14, 2021, virtual, pp. 1727-1738
Preview abstract
Distributed learning paradigms such as federated learning often involve transmission of model updates, or gradients, over a network, thereby avoiding transmission of private data. However, it is possible for sensitive information about the training data to be revealed from such gradients. Prior works have demonstrated that labels can be revealed analytically from the last layer of certain models (e.g., ResNet), or they can be reconstructed jointly with model inputs by using Gradients Matching [Zhu et al.] with additional knowledge about the current state of the model. In this work, we propose a method to discover the set of labels of training samples from only the gradient of the last layer and the id to label mapping. Our method is applicable to a wide variety of model architectures across multiple domains. We demonstrate the effectiveness of our method for model training in two domains - image classification, and automatic speech recognition. Furthermore, we show that existing reconstruction techniques improve their efficacy when used in conjunction with our method. Conversely, we demonstrate that gradient quantization and sparsification can significantly reduce the success of the attack.
View details
Understanding Unintended Memorization in Federated Learning
Third Workshop on Privacy in Natural Language Processing (PrivateNLP 2021) at 2021 Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL 2021)(2020)
Preview abstract
Recent works have shown that generative sequence models (e.g., language models) have a tendency to memorize rare or unique sequences in the training data. Since useful models are often trained on sensitive data, to ensure the privacy of the training data it is critical to identify and mitigate such unintended memorization. Federated Learning (FL) has emerged as a novel framework for large-scale distributed learning tasks. However, it differs in many aspects from the well-studied central learning setting where all the data is stored at the central server. In this paper, we initiate a formal study to understand the effect of different components of canonical FL on unintended memorization in trained models, comparing with the central learning setting. Our results show that several differing components of FL play an important role in reducing unintended memorization. Specifically, we observe that the clustering of data according to users---which happens by design in FL---has a significant effect in reducing such memorization, and using the method of Federated Averaging for training causes a further reduction. We also show that training with a strong user-level differential privacy guarantee results in models that exhibit the least amount of unintended memorization.
View details
Training Production Language Models without Memorizing User Data
Galen Andrew
(2020)
Preview abstract
This paper presents the first consumer-scale next-word prediction (NWP) model trained with Federated Learning (FL) while leveraging the Differentially Private Federated Averaging (DP-FedAvg) technique. There has been prior work on building practical FL infrastructure, including work demonstrating the feasibility of training language models on mobile devices using such infrastructure. It has also been shown (in simulations on a public corpus) that it is possible to train NWP models with user-level differential privacy using the DP-FedAvg algorithm. Nevertheless, training production-quality NWP models with DP-FedAvg in a real-world production environment on a heterogeneous fleet of mobile phones requires addressing numerous challenges. For instance, the coordinating central server has to keep track of the devices available at the start of each round and sample devices uniformly at random from them, while ensuring \emph{secrecy of the sample}, etc. Unlike all prior privacy-focused FL work of which we are aware, for the first time we demonstrate the deployment of a differentially private mechanism for the training of a production neural network in FL, as well as the instrumentation of the production training infrastructure to perform an end-to-end empirical measurement of unintended memorization.
View details
Preview abstract
We demonstrate that a character-level LSTM neural network is able to learn
out-of-vocabulary (OOV) words for the purpose of expanding the vocabulary of
a virtual keyboard for smartphones.
We train such a model using a distributed, on-device learning framework called federated learning.
High-frequency words can then be sampled from the generative model
by drawing from the joint posterior directly.
We study the feasibility of the approach in three different settings:
(1) using stochastic gradient descent, on an anonymized dataset of snippets of user content;
(2) using simulated federated learning, on a publicly available non-IID per-user dataset from a popular social networking website;
(3) using federated learning, on data hosted on user mobile devices.
The model is shown to achieve good recall and precision when compared to ground-truth OOV words in settings (1) and (2).
With (3) we demonstrate the practicality of this approach by showing that we can learn meaningful OOV words without exporting sensitive user data to servers.
View details
