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.) ?
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.
Catastrophic forgetting of connectionist neural networks is caused by the global sharing of parameters among all training examples. In this … (voir plus)study, we analyze parameter sharing under the conditional computation framework where the parameters of a neural network are conditioned on each input example. At one extreme, if each input example uses a disjoint set of parameters, there is no sharing of parameters thus no catastrophic forgetting. At the other extreme, if the parameters are the same for every example, it reduces to the conventional neural network. We then introduce a clipped version of maxout networks which lies in the middle, i.e. parameters are shared partially among examples. Based on the parameter sharing analysis, we can locate a limited set of examples that are interfered when learning a new example. We propose to perform rehearsal on this set to prevent forgetting, which is termed as conditional rehearsal. Finally, we demonstrate the effectiveness of the proposed method in an online non-stationary setup, where updates are made after each new example and the distribution of the received example shifts over time.
Near-Optimal Glimpse Sequences for Improved Hard Attention Neural Network Training
William Harvey
Michael Teng
Frank Wood
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.
We humans seem to have an innate understanding of the asymmetric progression of time, which we use to efficiently and safely perceive and ma… (voir plus)nipulate our environment. Drawing inspiration from that, we address the problem of learning an arrow of time in a Markov (Decision) Process. We illustrate how a learned arrow of time can capture meaningful information about the environment, which in turn can be used to measure reachability, detect side-effects and to obtain an intrinsic reward signal. We show empirical results on a selection of discrete and continuous environments, and demonstrate for a class of stochastic processes that the learned arrow of time agrees reasonably well with a known notion of an arrow of time given by the celebrated Jordan-Kinderlehrer-Otto result.
Model-based Reinforcement Learning approaches have the promise of being sample efficient. Much of the progress in learning dynamics models i… (voir plus)n RL has been made by learning models via supervised learning. But traditional model-based approaches lead to `compounding errors' when the model is unrolled step by step. Essentially, the state transitions that the learner predicts (by unrolling the model for multiple steps) and the state transitions that the learner experiences (by acting in the environment) may not be consistent. There is enough evidence that humans build a model of the environment, not only by observing the environment but also by interacting with the environment. Interaction with the environment allows humans to carry out experiments: taking actions that help uncover true causal relationships which can be used for building better dynamics models. Analogously, we would expect such interactions to be helpful for a learning agent while learning to model the environment dynamics. In this paper, we build upon this intuition by using an auxiliary cost function to ensure consistency between what the agent observes (by acting in the real world) and what it imagines (by acting in the `learned' world). We consider several tasks - Mujoco based control tasks and Atari games - and show that the proposed approach helps to train powerful policies and better dynamics models.
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.
In this note, we study the relationship between the variational gap and the variance of the (log) likelihood ratio. We show that the gap can… (voir plus) be upper bounded by some form of dispersion measure of the likelihood ratio, which suggests the bias of variational inference can be reduced by making the distribution of the likelihood ratio more concentrated, such as via averaging and variance reduction.
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.
