اشترك بالحزمة الذهبية واحصل على وصول غير محدود شمرا أكاديميا
تسجيل مستخدم جديدRevealing Distributional Vulnerability of Explicit Discriminators by Implicit Generators
68
0
0.0
(
0
)
اسأل ChatGPT حول البحث
ﻻ يوجد ملخص باللغة العربية
An explicit discriminator trained on observable in-distribution (ID) samples can make high-confidence prediction on out-of-distribution (OOD) samples due to its distributional vulnerability. This is primarily caused by the limited ID samples observable for training discriminators when OOD samples are unavailable. To address this issue, the state-of-the-art methods train the discriminator with OOD samples generated by general assumptions without considering the data and network characteristics. However, different network architectures and training ID datasets may cause diverse vulnerabilities, and the generated OOD samples thus usually misaddress the specific distributional vulnerability of the explicit discriminator. To reveal and patch the distributional vulnerabilities, we propose a novel method of textit{fine-tuning explicit discriminators by implicit generators} (FIG). According to the Shannon entropy, an explicit discriminator can construct its corresponding implicit generator to generate specific OOD samples without extra training costs. A Langevin Dynamic sampler then draws high-quality OOD samples from the generator to reveal the vulnerability. Finally, a regularizer, constructed according to the design principle of the implicit generator, patches the distributional vulnerability by encouraging those generated OOD samples with high entropy. Our experiments on four networks, four ID datasets and seven OOD datasets demonstrate that FIG achieves state-of-the-art OOD detection performance and maintains a competitive classification capability.
قيم البحث
اقرأ أيضاً
To improve the sample efficiency of policy-gradient based reinforcement learning algorithms, we propose implicit distributional actor-critic (IDAC) that consists of a distributional critic, built on two deep generator networks (DGNs), and a semi-impl
icit actor (SIA), powered by a flexible policy distribution. We adopt a distributional perspective on the discounted cumulative return and model it with a state-action-dependent implicit distribution, which is approximated by the DGNs that take state-action pairs and random noises as their input. Moreover, we use the SIA to provide a semi-implicit policy distribution, which mixes the policy parameters with a reparameterizable distribution that is not constrained by an analytic density function. In this way, the policys marginal distribution is implicit, providing the potential to model complex properties such as covariance structure and skewness, but its parameter and entropy can still be estimated. We incorporate these features with an off-policy algorithm framework to solve problems with continuous action space and compare IDAC with state-of-the-art algorithms on representative OpenAI Gym environments. We observe that IDAC outperforms these baselines in most tasks. Python code is provided.
Transformer models have recently attracted much interest from computer vision researchers and have since been successfully employed for several problems traditionally addressed with convolutional neural networks. At the same time, image synthesis usi
ng generative adversarial networks (GANs) has drastically improved over the last few years. The recently proposed TransGAN is the first GAN using only transformer-based architectures and achieves competitive results when compared to convolutional GANs. However, since transformers are data-hungry architectures, TransGAN requires data augmentation, an auxiliary super-resolution task during training, and a masking prior to guide the self-attention mechanism. In this paper, we study the combination of a transformer-based generator and convolutional discriminator and successfully remove the need of the aforementioned required design choices. We evaluate our approach by conducting a benchmark of well-known CNN discriminators, ablate the size of the transformer-based generator, and show that combining both architectural elements into a hybrid model leads to better results. Furthermore, we investigate the frequency spectrum properties of generated images and observe that our model retains the benefits of an attention based generator.
Combinatorial features are essential for the success of many commercial models. Manually crafting these features usually comes with high cost due to the variety, volume and velocity of raw data in web-scale systems. Factorization based models, which
measure interactions in terms of vector product, can learn patterns of combinatorial features automatically and generalize to unseen features as well. With the great success of deep neural networks (DNNs) in various fields, recently researchers have proposed several DNN-based factorization model to learn both low- and high-order feature interactions. Despite the powerful ability of learning an arbitrary function from data, plain DNNs generate feature interactions implicitly and at the bit-wise level. In this paper, we propose a novel Compressed Interaction Network (CIN), which aims to generate feature interactions in an explicit fashion and at the vector-wise level. We show that the CIN share some functionalities with convolutional neural networks (CNNs) and recurrent neural networks (RNNs). We further combine a CIN and a classical DNN into one unified model, and named this new model eXtreme Deep Factorization Machine (xDeepFM). On one hand, the xDeepFM is able to learn certain bounded-degree feature interactions explicitly; on the other hand, it can learn arbitrary low- and high-order feature interactions implicitly. We conduct comprehensive experiments on three real-world datasets. Our results demonstrate that xDeepFM outperforms state-of-the-art models. We have released the source code of xDeepFM at url{https://github.com/Leavingseason/xDeepFM}.
Deep ReLU networks trained with the square loss have been observed to perform well in classification tasks. We provide here a theoretical justification based on analysis of the associated gradient flow. We show that convergence to a solution with the
absolute minimum norm is expected when normalization techniques such as Batch Normalization (BN) or Weight Normalization (WN) are used together with Weight Decay (WD). The main property of the minimizers that bounds their expected error is the norm: we prove that among all the close-to-interpolating solutions, the ones associated with smaller Frobenius norms of the unnormalized weight matrices have better margin and better bounds on the expected classification error. With BN but in the absence of WD, the dynamical system is singular. Implicit dynamical regularization -- that is zero-initial conditions biasing the dynamics towards high margin solutions -- is also possible in the no-BN and no-WD case. The theory yields several predictions, including the role of BN and weight decay, aspects of Papyan, Han and Donohos Neural Collapse and the constraints induced by BN on the network weights.
A basic question in learning theory is to identify if two distributions are identical when we have access only to examples sampled from the distributions. This basic task is considered, for example, in the context of Generative Adversarial Networks (
GANs), where a discriminator is trained to distinguish between a real-life distribution and a synthetic distribution. % Classically, we use a hypothesis class $H$ and claim that the two distributions are distinct if for some $hin H$ the expected value on the two distributions is (significantly) different. Our starting point is the following fundamental problem: is having the hypothesis dependent on more than a single random example beneficial. To address this challenge we define $k$-ary based discriminators, which have a family of Boolean $k$-ary functions $mathcal{G}$. Each function $gin mathcal{G}$ naturally defines a hyper-graph, indicating whether a given hyper-edge exists. A function $gin mathcal{G}$ distinguishes between two distributions, if the expected value of $g$, on a $k$-tuple of i.i.d examples, on the two distributions is (significantly) different. We study the expressiveness of families of $k$-ary functions, compared to the classical hypothesis class $H$, which is $k=1$. We show a separation in expressiveness of $k+1$-ary versus $k$-ary functions. This demonstrate the great benefit of having $kgeq 2$ as distinguishers. For $kgeq 2$ we introduce a notion similar to the VC-dimension, and show that it controls the sample complexity. We proceed and provide upper and lower bounds as a function of our extended notion of VC-dimension.
الأسئلة المقترحة
سجل دخول لتتمكن من نشر تعليقات
التعليقات
جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
