Zero Shot Learning via Low-rank Embedded Semantic AutoEncoder

Abstract

Zero-shot learning (ZSL) has been widely researched and get successful in machine learning. Most existing ZSL methods aim to accurately recognize objects of unseen classes by learning a shared mapping from the feature space to a semantic space. However, such methods did not investigate in-depth whether the mapping can precisely reconstruct the original visual feature. Motivated by the fact that the data have low intrinsic dimensionality e.g. low-dimensional subspace. In this paper, we formulate a novel framework named Lowrank Embedded Semantic AutoEncoder (LESAE) to jointly seek a low-rank mapping to link visual features with their semantic representations. Taking the encoder-decoder paradigm, the encoder part aims to learn a low-rank mapping from the visual feature to the semantic space, while decoder part manages to reconstruct the original data with the learned mapping. In addition, a non-greedy iterative algorithm is adopted to solve our model. Extensive experiments on six benchmark datasets demonstrate its superiority over several state-ofthe-art algorithms.

Cooperative Coupled Generative Networks for Generalized Zero-Shot Learning

“…In TZSL, the search space is C u , i.e., Y = C u . Our algorithm is compared with 10 recently proposed baseline algorithms for TZSL task, including DEVISE [7], SJE [44], ALE [11], SYNC [18], SAE [13], DEM [14], GFZSL [45], LESAE [46], PSR [24], and RAS-GAN [26]. To show the effectiveness of the proposed, we compared the simulated results with 10 other algorithms.…”

Prototype Relaxation With Robust Principal Component Analysis for Zero Shot Learning

“…2. Concretely, besides the baseline results recorded in [25], we also provide the results of four other methods, including TVN [48], VZSL [49], LESAE [13] and LESD [11], among which LESAE and LESD are low rank based methods and most related to ours. From Tab.…”

“…The most relevant to ours are the low rank based methods [11], [13], [14], [23]. Ding et al in [11] assumed that the projection matrix from visual space to attribute space should have the characteristic of low rank, and exploited a constraint on singular values to solve the problem.…”

“…Ding et al in [11] assumed that the projection matrix from visual space to attribute space should have the characteristic of low rank, and exploited a constraint on singular values to solve the problem. Liu et al in [13] made the same assumption and apply the concept of SAE to constrain the projection matrix, which is solved by replacing the low rank constraint to a nuclear norm. Meng et al adopted the subspace learning strategy to ZSL by adopting Low Rank Representation (LRR) for visual features and utilizing locally linear subspaces to approximate the nonlinear manifold [23].…”

“…For example, Ding et al attached it to a attribute dictionary constraint and utilized an ensemble learn- ing strategy to optimize them [11]. Liu et al embedded it to a Semantic Autoencoder (SAE) [12], and employed a trace optimization method to solve it [13]. Niu et al tried to use the Internet browsed images as auxiliary weakly supervised information, and combined the low rank constraint to train a webly supervised model [14].…”