Le traitement du langage naturel à l'ère de l'IA générative
Rejoignez-nous à Mila en octobre pour un atelier de trois jour visant à explorer le potentiel de transformation des technologies langagières et leurs implications pour la société.
Ce programme est conçu pour fournir aux professionnel·le·s travaillant dans le domaine de la politique une compréhension fondamentale de la technologie de l'IA.
Nous utilisons des témoins pour analyser le trafic et l’utilisation de notre site web, afin de personnaliser votre expérience. Vous pouvez désactiver ces technologies à tout moment, mais cela peut restreindre certaines fonctionnalités du site. Consultez notre Politique de protection de la vie privée pour en savoir plus.
Paramètre des cookies
Vous pouvez activer et désactiver les types de cookies que vous souhaitez accepter. Cependant certains choix que vous ferez pourraient affecter les services proposés sur nos sites (ex : suggestions, annonces personnalisées, etc.).
Cookies essentiels
Ces cookies sont nécessaires au fonctionnement du site et ne peuvent être désactivés. (Toujours actif)
Cookies analyse
Acceptez-vous l'utilisation de cookies pour mesurer l'audience de nos sites ?
Multimedia Player
Acceptez-vous l'utilisation de cookies pour afficher et vous permettre de regarder les contenus vidéo hébergés par nos partenaires (YouTube, etc.) ?
Publications
On the interplay between noise and curvature and its effect on optimization and generalization
The speed at which one can minimize an expected loss using stochastic methods depends on two properties: the curvature of the loss and the v… (voir plus)ariance of the gradients. While most previous works focus on one or the other of these properties, we explore how their interaction affects optimization speed. Further, as the ultimate goal is good generalization performance, we clarify how both curvature and noise are relevant to properly estimate the generalization gap. Realizing that the limitations of some existing works stems from a confusion between these matrices, we also clarify the distinction between the Fisher matrix, the Hessian, and the covariance matrix of the gradients.
This work revisits the use of information criteria to characterize the generalization of deep learning models. In particular, we empirically… (voir plus) demonstrate the effectiveness of the Takeuchi information criterion (TIC), an extension of the Akaike information criterion (AIC) for misspecified models, in estimating the generalization gap, shedding light on why quantities such as the number of parameters cannot quantify generalization. The TIC depends on both the Hessian of the loss H and the covariance of the gradients C. By exploring the similarities and differences between these two matrices as well as the Fisher information matrix F, we study the interplay between noise and curvature in deep models. We also address the question of whether C is a reasonable approximation to F, as is commonly assumed.
This work revisits the use of information criteria to characterize the generalization of deep learning models. In particular, we empirically… (voir plus) demonstrate the effectiveness of the Takeuchi information criterion (TIC), an extension of the Akaike information criterion (AIC) for misspecified models, in estimating the generalization gap, shedding light on why quantities such as the number of parameters cannot quantify generalization. The TIC depends on both the Hessian of the loss H and the covariance of the gradients C. By exploring the similarities and differences between these two matrices as well as the Fisher information matrix F, we study the interplay between noise and curvature in deep models. We also address the question of whether C is a reasonable approximation to F, as is commonly assumed.
Social media sites are becoming a key factor in politics. These platforms are easy to manipulate for the purpose of distorting information s… (voir plus)pace to confuse and distract voters. Past works to identify disruptive patterns are mostly focused on analyzing the content of tweets. In this study, we jointly embed the information from both user posted content as well as a user's follower network, to detect groups of densely connected users in an unsupervised fashion. We then investigate these dense sub-blocks of users to flag anomalous behavior. In our experiments, we study the tweets related to the upcoming 2019 Canadian Elections, and observe a set of densely-connected users engaging in local politics in different provinces, and exhibiting troll-like behavior.
Hard visual attention is a promising approach to reduce the computational burden of modern computer vision methodologies. However, hard atte… (voir plus)ntion mechanisms can be difficult and slow to train, which is especially costly for applications like neural architecture search where multiple networks must be trained. We introduce a method to amortise the cost of training by generating an extra supervision signal for a subset of the training data. This supervision is in the form of sequences of ‘good’ locations to attend to for each image. We find that the best method to generate supervision sequences comes from framing hard attention for image classification as a Bayesian optimal experimental design (BOED) problem. From this perspective, the optimal locations to attend to are those which provide the greatest expected reduction in the entropy of the classification distribution. We introduce methodology from the BOED literature to approximate this optimal behaviour and generate ‘near-optimal’ supervision sequences. We then present a hard attention network training objective that makes use of these sequences and show that it allows faster training than prior work. We finally demonstrate the utility of faster hard attention training by incorporating supervision sequences in a neural architecture search, resulting in hard attention architectures which can outperform networks with access to the entire image.
Recent advances in variational inference enable the modelling of highly structured joint distributions, but are limited in their capacity to… (voir plus) scale to the high-dimensional setting of stochastic neural networks. This limitation motivates a need for scalable parameterizations of the noise generation process, in a manner that adequately captures the dependencies among the various parameters. In this work, we address this need and present the Kronecker Flow, a generalization of the Kronecker product to invertible mappings designed for stochastic neural networks. We apply our method to variational Bayesian neural networks on predictive tasks, PAC-Bayes generalization bound estimation, and approximate Thompson sampling in contextual bandits. In all setups, our methods prove to be competitive with existing methods and better than the baselines.
We consider a communication system where multiple unknown channels are available for transmission. Each channel is a channel with state whic… (voir plus)h evolves in a Markov manner. The transmitter has to select L channels to use and also decide the resources (e.g., power, rate, etc.) to use for each of the selected channels. It observes the state of the channels it uses and receives no feedback on the state of the other channels. We model this problem as a partially observable Markov decision process and obtain a simplified belief state. We show that the optimal resource allocation policy can be identified in closed form. Once the optimal resource allocation policy is fixed, choosing the channel scheduling policy may be viewed as a restless bandit. We present an efficient algorithm to check indexability and compute the Whittle index for each channel. When the model is indexable, the Whittle index policy, which transmits over the L channels with the smallest Whittle indices, is an attractive heuristic policy.
Neural abstractive summarizers generate summary texts using a language model conditioned on the input source text, and have recently achieve… (voir plus)d high ROUGE scores on benchmark summarization datasets. We investigate how they achieve this performance with respect to human-written gold-standard abstracts, and whether the systems are able to understand deeper syntactic and semantic structures. We generate a set of contrastive summaries which are perturbed, deficient versions of human-written summaries, and test whether existing neural summarizers score them more highly than the human-written summaries. We analyze their performance on different datasets and find that these systems fail to understand the source text, in a majority of the cases.
2019-06-01
North American Chapter of the Association for Computational Linguistics (publié)
Existing controllable text generation systems rely on annotated attributes, which greatly limits their capabilities and applications. In thi… (voir plus)s work, we make the first successful attempt to use VAEs to achieve controllable text generation without supervision. We do so by decomposing the latent space of the VAE into two parts: one incorporates structural constraints to capture dominant global variations implicitly present in the data, e.g., sentiment or topic; the other is unstructured and is used for the reconstruction of the source sentences. With the enforced structural constraint, the underlying global variations will be discovered and disentangled during the training of the VAE. The structural constraint also provides a natural recipe for mitigating posterior collapse for the structured part, which cannot be fully resolved by the existing techniques. On the task of text style transfer, our unsupervised approach achieves significantly better performance than previous supervised approaches. By showcasing generation with finer-grained control including Cards-Against-Humanity-style topic transitions within a sentence, we demonstrate that our model can perform controlled text generation in a more flexible way than existing methods.
Contributors to open source software (OSS) communities assume diverse roles to take different responsibilities. One major limitation of the … (voir plus)current OSS tools and platforms is that they provide a uniform user interface regardless of the activities performed by the various types of contributors. This paper serves as a non-trivial first step towards resolving this challenge by demonstrating a methodology and establishing knowledge to understand how the contributors' roles and their dynamics, reflected in the activities contributors perform, are exhibited in OSS communities. Based on an analysis of user action data from 29 GitHub projects, we extracted six activities that distinguished four Active roles and five Supporting roles of OSS contributors, as well as patterns in role changes. Through the lens of the Activity Theory, these findings provided rich design guidelines for OSS tools to support diverse contributor roles.
2019-05-27
2019 IEEE/ACM 12th International Workshop on Cooperative and Human Aspects of Software Engineering (CHASE) (publié)
Fixing Bias in Reconstruction-based Anomaly Detection with Lipschitz Discriminators
Alexander Tong
Guy Wolf
Smita Krishnaswamy
Anomaly detection is of great interest in fields where abnormalities need to be identified and corrected (e.g., medicine and finance). Deep … (voir plus)learning methods for this task often rely on autoencoder reconstruction error, sometimes in conjunction with other penalties. We show that this approach exhibits intrinsic biases that lead to undesirable results. Reconstruction-based methods can sometimes show low error on simple-to-reconstruct points that are not part of the training data, for example the all black image. Instead, we introduce a new unsupervised Lipschitz anomaly discriminator (LAD) that does not suffer from these biases. Our anomaly discriminator is trained, similar to the discriminator of a GAN, to detect the difference between the training data and corruptions of the training data. We show that this procedure successfully detects unseen anomalies with guarantees on those that have a certain Wasserstein distance from the data or corrupted training set. These additions allow us to show improved performance on MNIST, CIFAR10, and health record data. Further, LAD does not require decoding back to the original data space, which makes anomaly detection possible in domains where it is difficult to define a decoder, such as in irregular graph structured data. Empirically, we show this framework leads to improved performance on image, health record, and graph data.