Abstract. Multi-view learning techniques are necessary when data is described by multiple distinct feature sets because single-view learning algorithms tend to overfit on these high-dimensional data. Prior successful approaches followed either consensus or complementary principles. Recent work has focused on learning both the shared and private latent spaces of views in order to take advantage of both principles. However, these methods can not ensure that the latent spaces are strictly independent through encouraging the orthogonality in their objective functions. Also little work has explored representation learning techniques for multiview learning. In this paper, we use the denoising autoencoder to learn shared and private latent spaces, with orthogonal constraints -disconnecting every private latent space from the remaining views. Instead of computationally expensive optimization, we adapt the backpropagation algorithm to train our model.

Abstract. Nonnegative Matrix Factorization (NMF) has received considerable attention due to its psychological and physiological interpretation of naturally occurring data whose representation may be partsbased in the human brain. However, when labeled and unlabeled images are sampled from different distributions, they may be quantized into different basis vector space and represented in different coding vector space, which may lead to low representation fidelity. In this paper, we investigate how to extend NMF to cross-domain scenario. We accomplish this goal through TNMF -a novel semi-supervised transfer learning approach. Specifically, we aim to minimize the distribution divergence between labeled and unlabeled images, and incorporate this criterion into the objective function of NMF to construct new robust representations. Experiments show that TNMF outperforms state-of-the-art methods on real datasets.

Wearable devices such as Google Glass are receiving increasing attention and look set to become part of our technical landscape over the next few years. At the same time, lifelogging is a topic that is growing in popularity with a host of new devices on the market that visually capture life experience in an automated manner. In this paper, we describe a visual lifelogging solution for Google Glass that is designed to capture life experience in rich visual detail, yet maintain the privacy of unknown bystanders.We present the approach called negative face blurring and evaluate it on a collection of lifelogging data of around nine thousand pictures from Google Glass.

In an era of model and data proliferation in machine learning/AI especially marked by the rapid advancement of open-sourced technologies, there arises a critical need for standardized consistent documentation. Our work addresses the information incompleteness in current human-generated model and data cards. We propose an automated generation approach using Large Language Models (LLMs). Our key contributions include the establishment of CARDBENCH, a comprehensive dataset aggregated from over 4.8k model cards and 1.4k data cards, coupled with the development of the CARDGEN pipeline comprising a two-step retrieval process. Our approach exhibits enhanced completeness, objectivity, and faithfulness in generated model and data cards, a significant step in responsible AI documentation practices ensuring better accountability and traceability. 1

Demographic attributes prediction is fundamental and important in many applications in real world, such as: recommendation, personalized search and behavior targeting. Although a variety of subjects are involved with demographic attributes prediction, e.g. there are requirements to recognize and predict demography from psychology, but the traditional approach is dynamic modeling on specified field and distinctive datasets. However, dynamic modeling takes researchers a lot of time and energy, even if it is done, no one has an idea how good or how bad it is. To tackle the problems mentioned above, a framework is proposed in this chapter to predict using classifiers as core part, which consists of three main components: data processing, predicting using classifiers and prediction adjustments. The component of data processing performs to clean and format data. The first step is extracting relatively independent data from complicated original dataset. In the next step, the extracted data goes through different paths based on their types. And at the last step, all the data will be transformed into a demographic attributes matrix. To fulfill prediction, the demographic attributes matrix is taken as the input of classifiers, and the testing