Suffering from the extreme training data imbalance between seen and unseen classes, most of existing state-of-theart approaches fail to achieve satisfactory results for the challenging generalized zero-shot learning task. To circumvent the need for labeled examples of unseen classes, we propose a novel generative adversarial network (GAN) that synthesizes CNN features conditioned on class-level...

Part of the appeal of Visual Question Answering (VQA) is its promise to answer new questions about previously unseen images. Most current methods demand training questions that illustrate every possible concept, and will therefore never achieve this capability, since the volume of required training data would be prohibitive. Answering general questions about images requires methods capable of Z...

One of the main challenges in learning fine-grained visual categories is gathering training images. Recent work in Zero-Shot Learning (ZSL) circumvents this challenge by describing categories via attributes or text. However, not all visual concepts, e.g ., two people dancing, are easily amenable to such descriptions. In this paper, we propose a new modality for ZSL using visual abstraction to l...

We present a generative framework for zero-shot action recognition where some of the possible action classes do not occur in the training data. Our approach is based on modeling each action class using a probability distribution whose parameters are functions of the attribute vector representing that action class. In particular, we assume that the distribution parameters for any action class in...

We introduce 0 | • (“zero hand-grenade”) as a sharp for an inner model with a proper class of strong cardinals. We prove the existence of the core model K in the theory “ZFC + 0 | • doesn’t exist,” and apply this to obtain an equiconsistency result.

Current state-of-the-art approaches in domain adaptation and fusion show promising results with either labeled or unlabeled task-relevant target-domain training data. However, the fact that the task-relevant target-domain training data can be unavailable is often ignored by the prior works. To tackle this issue, instead of using the task-relevant target-domain training data, we propose zeroshot...

In this paper, we are interested in the few-shot learning problem. In particular, we focus on a challenging scenario where the number of categories is large and the number of examples per novel category is very limited, i.e. 1, 2, or 3. Motivated by the close relationship between the parameters and the activations in a neural network associated with the same category, we propose a novel method ...

the notion of a bead metric space is defined as a nice generalization of the uniformly convex normed space such as $cat(0)$ space, where the curvature is bounded from above by zero. in fact, the bead spaces themselves can be considered in particular as natural extensions of convex sets in uniformly convex spaces and normed bead spaces are identical with uniformly convex spaces. in this paper, w...

When building spoken dialogue systems for a new domain, a major bottleneck is developing a spoken language understanding (SLU) module that handles the new domain’s terminology and semantic concepts. We propose a statistical SLU model that generalises to both previously unseen input words and previously unseen output classes by leveraging unlabelled data. After mapping the utterance into a vecto...

smooth coefficients aα(x), L+ be a formally adjoint differential operation. Let L0, L0 be minimal operators (i.e., for example, D(L0) is the clozure of C∞ 0 (Ω) in the norm of the graph ‖u‖L = ‖u‖L2(Ω)+‖Lu‖L2(Ω)), and L, L+ be maximal expansions of L,L+ in the space L2(Ω) respectively (i.e. L = (L0 ) ∗, L+ = (L0)∗), L̃ = L|D(L̃) where D(L̃) is the clozure of C∞(Ω̄) in the norm of the graph ‖u‖L and...