Google has disclosed ongoing efforts to improve its search engine using its machine learning software called Multitask Unified Model (MUM).

Pre-trained representation is one of the key elements in the success of modern deep learning. However, existing works on continual learning methods have mostly focused on learning models incrementally from scratch. In this paper, we explore an alternative framework to incremental learning where we continually fine-tune the model from a pre-trained representation. Our method takes advantage of linearization technique of a pre-trained neural network for simple and effective continual learning. We show that this allows us to design a linear model where quadratic parameter regularization method is placed as the optimal continual learning policy, and at the same time enjoying the high performance of neural networks. We also show that the proposed algorithm enables parameter regularization methods to be applied to class-incremental problems. Additionally, we provide a theoretical reason why the existing parameter-space regularization algorithms such as EWC underperform on neural networks trained with cross-entropy loss. We show that the proposed method can prevent forgetting while achieving high continual fine-tuning performance on image classification tasks. To show that our method can be applied to general continual learning settings, we evaluate our method in data-incremental, task-incremental, and class-incremental learning problems.

In this paper, we introduce a structure-based neural network architecture, namely RC-Struct, for MIMO-OFDM symbol detection. The RC-Struct exploits the temporal structure of the MIMO-OFDM signals through reservoir computing (RC). A binary classifier leverages the repetitive constellation structure in the system to perform multi-class detection. The incorporation of RC allows the RC-Struct to be learned in a purely online fashion with extremely limited pilot symbols in each OFDM subframe. The binary classifier enables the efficient utilization of the precious online training symbols and allows an easy extension to high-order modulations without a substantial increase in complexity. Experiments show that the introduced RC-Struct outperforms both the conventional model-based symbol detection approaches and the state-of-the-art learning-based strategies in terms of bit error rate (BER). The advantages of RC-Struct over existing methods become more significant when rank and link adaptation are adopted. The introduced RC-Struct sheds light on combining communication domain knowledge and learning-based receive processing for 5G/5G-Advanced and Beyond.

Abstract--Real world data often exhibits a long-tailed and open-ended (i.e. with unseen classes) distribution. A practical recognition system must balance between majority (head) and minority (tail) classes, generalize across the distribution, and acknowledge novelty upon the instances of unseen classes (open classes). We define Open Long-Tailed Recognition++ (OLTR++) as learning from such naturally distributed data and optimizing for the classification accuracy over a balanced test set which includes both known and open classes. OLTR++ handles imbalanced classification, few-shot learning, open-set recognition, and active learning in one integrated algorithm, whereas existing classification approaches often focus only on one or two aspects and deliver poorly over the entire spectrum. The key challenges are: 1) how to share visual knowledge between head and tail classes, 2) how to reduce confusion between tail and open classes, and 3) how to actively explore open classes with learned knowledge. Our algorithm, OLTR++, maps images to a feature space such that visual concepts can relate to each other through a memory association mechanism and a learned metric (dynamic meta-embedding) that both respects the closed world classification of seen classes and acknowledges the novelty of open classes. Additionally, we propose an active learning scheme based on visual memory, which learns to recognize open classes in a data-efficient manner for future expansions. On three large-scale open long-tailed datasets we curated from ImageNet (object-centric), Places (scene-centric), and MS1M (face-centric) data, as well as three standard benchmarks (CIFAR-10-LT, CIFAR-100-LT, and iNaturalist-18), our approach, as a unified framework, consistently demonstrates competitive performance. Notably, our approach also shows strong potential for the active exploration of open classes and the fairness analysis of minority groups.

Lecture slide presentations, a sequence of pages that contain text and figures accompanied by speech, are constructed and presented carefully in order to optimally transfer knowledge to students. Previous studies in multimedia and psychology attribute the effectiveness of lecture presentations to their multimodal nature. As a step toward developing AI to aid in student learning as intelligent teacher assistants, we introduce the Multimodal Lecture Presentations dataset as a large-scale benchmark testing the capabilities of machine learning models in multimodal understanding of educational content. Our dataset contains aligned slides and spoken language, for 180+ hours of video and 9000+ slides, with 10 lecturers from various subjects (e.g., computer science, dentistry, biology). We introduce two research tasks which are designed as stepping stones towards AI agents that can explain (automatically captioning a lecture presentation) and illustrate (synthesizing visual figures to accompany spoken explanations) educational content. We provide manual annotations to help implement these two research tasks and evaluate state-of-the-art models on them. Comparing baselines and human student performances, we find that current models struggle in (1) weak crossmodal alignment between slides and spoken text, (2) learning novel visual mediums, (3) technical language, and (4) long-range sequences. Towards addressing this issue, we also introduce PolyViLT, a multimodal transformer trained with a multi-instance learning loss that is more effective than current approaches. We conclude by shedding light on the challenges and opportunities in multimodal understanding of educational presentations.

Phase retrieval (PR), a long-established challenge for recovering a complex-valued signal from its Fourier intensity-only measurements, has attracted considerable attention due to its widespread applications in digital imaging. Recently, deep learning-based approaches were developed that achieved some success in single-shot PR. These approaches require a single Fourier intensity measurement without the need to impose any additional constraints on the measured data. Nevertheless, vanilla deep neural networks (DNN) do not give good performance due to the substantial disparity between the input and output domains of the PR problems. Physics-informed approaches try to incorporate the Fourier intensity measurements into an iterative approach to increase the reconstruction accuracy. It, however, requires a lengthy computation process, and the accuracy still cannot be guaranteed. Besides, many of these approaches work on simulation data that ignore some common problems such as saturation and quantization errors in practical optical PR systems. In this paper, a novel physics-driven multi-scale DNN structure dubbed PPRNet is proposed. Similar to other deep learning-based PR methods, PPRNet requires only a single Fourier intensity measurement. It is physics-driven that the network is guided to follow the Fourier intensity measurement at different scales to enhance the reconstruction accuracy. PPRNet has a feedforward structure and can be end-to-end trained. Thus, it is much faster and more accurate than the traditional physics-driven PR approaches. Extensive simulations and experiments on a practical optical platform were conducted. The results demonstrate the superiority and practicality of the proposed PPRNet over the traditional learning-based PR methods.

Prognostics and Health Management (PHM) is an emerging field which has received much attention from the manufacturing industry because of the benefits and efficiencies it brings to the table. And Remaining Useful Life (RUL) prediction is at the heart of any PHM system. Most recent data-driven research demand substantial volumes of labelled training data before a performant model can be trained under the supervised learning paradigm. This is where Transfer Learning (TL) and Domain Adaptation (DA) methods step in and make it possible for us to generalize a supervised model to other domains with different data distributions with no labelled data. In this paper, we propose \textit{LAMA-Net}, an encoder-decoder based model (Transformer) with an induced bottleneck, Latent Alignment using Maximum Mean Discrepancy (MMD) and manifold learning is proposed to tackle the problem of Unsupervised Homogeneous Domain Adaptation for RUL prediction. \textit{LAMA-Net} is validated using the C-MAPSS Turbofan Engine dataset by NASA and compared against other state-of-the-art techniques for DA. The results suggest that the proposed method offers a promising approach to perform domain adaptation in RUL prediction. Code will be made available once the paper comes out of review.

The description of "artificial intelligence in high school" may conjure up a science fiction novel where robots stand around chatting at their lockers. The reality, at Seckinger High School in Gwinnett County, Ga., looks more like this: A social studies teacher pauses a lesson on the spread of cholera in the 19th century to discuss how data scientists use AI tools today to track diseases. A math class full of English-language learners uses machine learning to identify linear and non-linear shapes. The simplest explanation of this technology is that it trains a machine to do tasks that simulate some of what the human brain can do. That means it can learn to do things like recognize faces and voices (helpful for radiology, security, and more), understand natural language, and even make recommendations.

As machine learning is applied to an increasingly large number of domains, the need for an effective way to explain its predictions grows apace. In the domain of child welfare screening, machine learning offers a promising method of consolidating the large amount of data that screeners must look at, potentially improving the outcomes for children reported to child welfare departments. Interviews and case-studies suggest that adding an explanation alongside the model prediction may result in better outcomes, but it is not obvious what kind of explanation would be most useful in this context. Through a series of interviews and user studies, we developed Sibyl, a machine learning explanation dashboard specifically designed to aid child welfare screeners' decision making. When testing Sibyl, we evaluated four different explanation types, and based on this evaluation, decided a local feature contribution approach was most useful to screeners.

Machine learning can make your applications faster and more intelligent. You can analyze customer data such as voice and text input, images, and video, and take action without human intervention. Google Cloud Platform (GCP) offers a competitive set of machine learning services for nearly every type of architecture, including serverless computing, containers, and virtual machines. Learn how to design your own machine learning solutions using GCP, in this introductory course with instructor Lynn Langit. Lynn shows how to identify your requirements and map them to services such as the GCP machine learning APIs--Cloud Vision, Cloud Speech-to-Text, Cloud Video Intelligence, and more--and GCP AutoML, which puts the same APIs behind an easy-to-use interface.