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.) ?
Generative Adversarial Networks (GANs) are powerful generative models, but suffer from training instability. The recently proposed Wasserste… (voir plus)in GAN (WGAN) makes progress toward stable training of GANs, but sometimes can still generate only low-quality samples or fail to converge. We find that these problems are often due to the use of weight clipping in WGAN to enforce a Lipschitz constraint on the critic, which can lead to undesired behavior. We propose an alternative to clipping weights: penalize the norm of gradient of the critic with respect to its input. Our proposed method performs better than standard WGAN and enables stable training of a wide variety of GAN architectures with almost no hyperparameter tuning, including 101-layer ResNets and language models over discrete data. We also achieve high quality generations on CIFAR-10 and LSUN bedrooms.
It is commonly assumed that language refers to high-level visual concepts while leaving low-level visual processing unaffected. This view do… (voir plus)minates the current literature in computational models for language-vision tasks, where visual and linguistic input are mostly processed independently before being fused into a single representation. In this paper, we deviate from this classic pipeline and propose to modulate the \emph{entire visual processing} by linguistic input. Specifically, we condition the batch normalization parameters of a pretrained residual network (ResNet) on a language embedding. This approach, which we call MOdulated RESnet (\MRN), significantly improves strong baselines on two visual question answering tasks. Our ablation study shows that modulating from the early stages of the visual processing is beneficial.
We consider the problem of estimating multiple related functions computed by weighted automata~(WFA). We first present a natural notion of r… (voir plus)elatedness between WFAs by considering to which extent several WFAs can share a common underlying representation. We then introduce the model of vector-valued WFA which conveniently helps us formalize this notion of relatedness. Finally, we propose a spectral learning algorithm for vector-valued WFAs to tackle the multitask learning problem. By jointly learning multiple tasks in the form of a vector-valued WFA, our algorithm enforces the discovery of a representation space shared between tasks. The benefits of the proposed multitask approach are theoretically motivated and showcased through experiments on both synthetic and real world datasets.
Piecewise Latent Variables for Neural Variational Text Processing
Advances in neural variational inference have facilitated the learning of powerful directed graphical models with continuous latent variable… (voir plus)s, such as variational autoencoders. The hope is that such models will learn to represent rich, multi-modal latent factors in real-world data, such as natural language text. However, current models often assume simplistic priors on the latent variables - such as the uni-modal Gaussian distribution - which are incapable of representing complex latent factors efficiently. To overcome this restriction, we propose the simple, but highly flexible, piecewise constant distribution. This distribution has the capacity to represent an exponential number of modes of a latent target distribution, while remaining mathematically tractable. Our results demonstrate that incorporating this new latent distribution into different models yields substantial improvements in natural language processing tasks such as document modeling and natural language generation for dialogue.
2017-01-01
Conference on Empirical Methods in Natural Language Processing (publié)
Natural image modeling is a landmark challenge of unsupervised learning. Variational Autoencoders (VAEs) learn a useful latent representatio… (voir plus)n and model global structure well but have difficulty capturing small details. PixelCNN models details very well, but lacks a latent code and is difficult to scale for capturing large structures. We present PixelVAE, a VAE model with an autoregressive decoder based on PixelCNN. Our model requires very few expensive autoregressive layers compared to PixelCNN and learns latent codes that are more compressed than a standard VAE while still capturing most non-trivial structure. Finally, we extend our model to a hierarchy of latent variables at different scales. Our model achieves state-of-the-art performance on binarized MNIST, competitive performance on 64 × 64 ImageNet, and high-quality samples on the LSUN bedrooms dataset.
Many real world applications include information on both attributes of individual entities as well as relations between them, while there ex… (voir plus)ists an interplay between these attributes and relations. For example, in a typical social network, the similarity of individuals’ characteristics motivates them to form relations, a.k.a. social selection; whereas the characteristics of individuals may be affected by the characteristics of their relations, a.k.a. social influence. We can measure proclivity in networks by quantifying the correlation of nodal attributes and the structure [1]. Here, we are interested in a more fundamental study, to extend the basic statistics defined for graphs and draw parallels for the attributed graphs. More formally, an attributed graph is denoted by (A,X); where An×n is the adjacency matrix and encodes the relationships between the n nodes, and Xn×k is the attributes matrix –each row shows the feature vector of the corresponding node. Degree of a node encodes the number of its neighbors, computed as ki = ∑ j Aij . We can extend this notion to networks with binary attributes to the number of neighbors which share a particular attribute x, i.e. ki(x) = ∑ j Aijδ(Xj , x); where δ(Xj , x) = 1 iff node j has attribute x. Similar to the simple graphs, where the degree distribution is studied and showed to be heavy tail, here we can look at: 1) the degree distributions per attribute, 2) the joint probability distribution of any pair of attributes. Moreover, if we assume A(x1, x2) is the induced subgraph (or masked matrix of edges) with endpoints of values (x1, x2), i.e., A(x1, x2) = Aijδ(Xi, x1)δ(Xj , x2), then we can study and compare these distributions for the induced subgraph per each pair of attribute values. For example, Figure 1 shows the same general trend in the distribution of the original graph and the three possible induced subgraph.
We propose a reparameterization of LSTM that brings the benefits of batch normalization to recurrent neural networks. Whereas previous works… (voir plus) only apply batch normalization to the input-to-hidden transformation of RNNs, we demonstrate that it is both possible and beneficial to batch-normalize the hidden-to-hidden transition, thereby reducing internal covariate shift between time steps.
We evaluate our proposal on various sequential problems such as sequence classification, language modeling and question answering. Our empirical results show that our batch-normalized LSTM consistently leads to faster convergence and improved generalization.
In this paper we propose a novel model for unconditional audio generation task that generates one audio sample at a time. We show that our m… (voir plus)odel which profits from combining memory-less modules, namely autoregressive multilayer perceptron, and stateful recurrent neural networks in a hierarchical structure is de facto powerful to capture the underlying sources of variations in temporal domain for very long time on three datasets of different nature. Human evaluation on the generated samples indicate that our model is preferred over competing models. We also show how each component of the model contributes to the exhibited performance.
We propose a new sequential algorithm for Sampling Importance Resampling. The algorithm serves as a solution to expensive evaluation of impo… (voir plus)rtance weight, and can be interpreted as stochastically and iteratively refining the particles by correcting them towards the target distribution as pool size increases. We apply this algorithm to variational inference with Importance Weighted Lower Bound and propose a memory-scalable training procedure 1 that implicitly improves the variational proposal. 1 Sequentializing Sampling Importance Resampling 1.1 Sampling Importance Resampling Given an unnormalized target distribution p̃(x) and proposal distribution q(x), the Sampling Importance Resampling (SIR) proceeds as follows: 1. draw xi for 1 ≤ i ≤ n from q(x) 2. calculate the importance weight wi = p̃(xi) q(xi) 3. calculate the normalized importance weight w̄i = wi ∑ i wi 4. draw index variable yj ∼ mul(w̄1, ..., w̄n) for 1 ≤ j ≤ m The density of the set of resampled particles xy1 , ..., xym should resemble the pdf of the target distribution, and the new samples will be approximately distributed by p(x) (Bishop, 2007). On average, the samples can be improved by increasing the pool size n, and becomes corrected when n→∞. The procedure is visualized in Figure 1a. 1.2 SeqSIR The above procedure can be combined with the idea of reservoir sampling, so that we need not evaluate all n samples at the same time, which will be an issue when n is large or when evaluation of a sample (i.e. computation of wi) is expensive. The intuition is to keep a running sum of the importance weights while we evaluate the pool samples sequentially, and then decide to keep the old sample or replace it with the new one based on the ratio of the new sample’s importance weight to the running sum. This is what we call Sequentialized Sampling Importance Resampling (SEQSIR), which is summarized in Algorithm 1. See Figure 1b for illustration. Note that density and importance weight are computed on log scale to deal with numerical instability, and log-sum-exp operation (LSE) is used in place of addition to calculate the running sum of See https://github.com/CW-Huang/SeqIWAE for implementation. Second workshop on Bayesian Deep Learning (NIPS 2017), Long Beach, CA, USA. Algorithm 1 Sequentialized Sampling Importance Resampling and Stochastic Iterative Refinement procedure SEQSIR ( logp, logq . unnormalized target density function and proposal density function ss . n samples to be evaluated ) A←−∞ . initialize accumulated sum of importance weight on log scale s_old← 0 . initialize sample n← len([s1,...,sn]) for i=1,...,n do s_new = ss[i] A, s_old← STOCHREFINE(logp, logq, A, s_old, s_new) return s_old procedure STOCHREFINE ( logp, logq . unnormalized target density function and proposal density function A . accumulated sum of importance weight on log scale s_old, s_new . old and new samples ) w_new← logp(s_new) logq(s_new) A← LSE(A, w_new) u← unif(0,1) if w_new A >= log u then return A, s_new else return A, s_old
