Established experimental procedures for one-shot machine learning do not test the ability to learn or remember specific instances of classes, a key feature of animal intelligence. Distinguishing specific instances is necessary for many real-world tasks, such as remembering which cup belongs to you. Generalisation within classes conflicts with the ability to separate instances of classes, making it difficult to achieve both capabilities within a single architecture. We propose an extension to the standard Omniglot classification-generalisation framework that additionally tests the ability to distinguish specific instances after one exposure and introduces noise and occlusion corruption. Learning is defined as an ability to classify as well as recall training samples. Complementary Learning Systems (CLS) is a popular model of mammalian brain regions believed to play a crucial role in learning from a single exposure to a stimulus. We created an artificial neural network implementation of CLS and applied it to the extended Omniglot benchmark. Our unsupervised model demonstrates comparable performance to existing supervised ANNs on the Omniglot classification task (requiring generalisation), without the need for domain-specific inductive biases. On the extended Omniglot instance-recognition task, the same model also demonstrates significantly better performance than a baseline nearest-neighbour approach, given partial occlusion and noise.

Question answering is an effective method for obtaining information from knowledge bases (KB). In this paper, we propose the Neural-Symbolic Complex Question Answering (NS-CQA) model, a data-efficient reinforcement learning framework for complex question answering by using only a modest number of training samples. Our framework consists of a neural generator and a symbolic executor that, respectively, transforms a natural-language question into a sequence of primitive actions, and executes them over the knowledge base to compute the answer. We carefully formulate a set of primitive symbolic actions that allows us to not only simplify our neural network design but also accelerate model convergence. To reduce search space, we employ the copy and masking mechanisms in our encoder-decoder architecture to drastically reduce the decoder output vocabulary and improve model generalizability. We equip our model with a memory buffer that stores high-reward promising programs. Besides, we propose an adaptive reward function. By comparing the generated trial with the trials stored in the memory buffer, we derive the curriculum-guided reward bonus, i.e., the proximity and the novelty. To mitigate the sparse reward problem, we combine the adaptive reward and the reward bonus, reshaping the sparse reward into dense feedback. Also, we encourage the model to generate new trials to avoid imitating the spurious trials while making the model remember the past high-reward trials to improve data efficiency. Our NS-CQA model is evaluated on two datasets: CQA, a recent large-scale complex question answering dataset, and WebQuestionsSP, a multi-hop question answering dataset. On both datasets, our model outperforms the state-of-the-art models. Notably, on CQA, NS-CQA performs well on questions with higher complexity, while only using approximately 1% of the total training samples.

Complex question-answering (CQA) involves answering complex natural-language questions on a knowledge base (KB). However, the conventional neural program induction (NPI) approach exhibits uneven performance when the questions have different types, harboring inherently different characteristics, e.g., difficulty level. This paper proposes a meta-reinforcement learning approach to program induction in CQA to tackle the potential distributional bias in questions. Our method quickly and effectively adapts the meta-learned programmer to new questions based on the most similar questions retrieved from the training data. The meta-learned policy is then used to learn a good programming policy, utilizing the trial trajectories and their rewards for similar questions in the support set. Our method achieves state-of-the-art performance on the CQA dataset (Saha et al., 2018) while using only five trial trajectories for the top-5 retrieved questions in each support set, and metatraining on tasks constructed from only 1% of the training set. We have released our code at https://github.com/DevinJake/MRL-CQA.

A compelling approach to complex question answering is to convert the question to a sequence of actions, which can then be executed on the knowledge base to yield the answer, aka the programmer-interpreter approach. Use similar training questions to the test question, meta-learning enables the programmer to adapt to unseen questions to tackle potential distributional biases quickly. However, this comes at the cost of manually labeling similar questions to learn a retrieval model, which is tedious and expensive. In this paper, we present a novel method that automatically learns a retrieval model alternately with the programmer from weak supervision, i.e., the system's performance with respect to the produced answers. To the best of our knowledge, this is the first attempt to train the retrieval model with the programmer jointly. Our system leads to state-of-the-art performance on a large-scale task for complex question answering over knowledge bases. We have released our code at https://github.com/DevinJake/MARL.

Edge intelligence is an emerging network architecture that integrates sensing, communication, computing components, and supports various machine learning applications, where a fundamental communication question is: how to allocate the limited wireless resources (such as time, energy) to the simultaneous model training of heterogeneous learning tasks? Existing methods ignore two important facts: 1) different models have heterogeneous demands on training data; 2) there is a mismatch between the simulated environment and the real-world environment. As a result, they could lead to low learning performance in practice. This paper proposes the learning centric wireless resource allocation (LCWRA) scheme that maximizes the worst learning performance of multiple classification tasks. Analysis shows that the optimal transmission time has an inverse power relationship with respect to the classification error. Finally, both simulation and experimental results are provided to verify the performance of the proposed LCWRA scheme and its robustness in real implementation.

Word embeddings are usually derived from corpora containing text from many individuals, thus leading to general purpose representations rather than individually personalized representations. While personalized embeddings can be useful to improve language model performance and other language processing tasks, they can only be computed for people with a large amount of longitudinal data, which is not the case for new users. We propose a new form of personalized word embeddings that use demographic-specific word representations derived compositionally from full or partial demographic information for a user (i.e., gender, age, location, religion). We show that the resulting demographic-aware word representations outperform generic word representations on two tasks for English: language modeling and word associations. We further explore the trade-off between the number of available attributes and their relative effectiveness and discuss the ethical implications of using them.

The graph represents a network of 2,067 Twitter users whose tweets in the requested range contained "#cloudcomputing", or who were replied to or mentioned in those tweets. The network was obtained from the NodeXL Graph Server on Monday, 26 October 2020 at 12:02 UTC. The requested start date was Monday, 26 October 2020 at 00:01 UTC and the maximum number of days (going backward) was 14. The maximum number of tweets collected was 7,500. The tweets in the network were tweeted over the 3-day, 9-hour, 0-minute period from Thursday, 22 October 2020 at 14:58 UTC to Sunday, 25 October 2020 at 23:58 UTC.

In 1953, US senators grilled General Motors CEO Charles "Engine Charlie" Wilson about his large GM shareholdings: Would they cloud his decision making if he became the US secretary of defense and the interests of General Motors and the United States diverged? Wilson said that he would always put US interests first but that he could not imagine such a divergence taking place, because, "for years I thought what was good for our country was good for General Motors, and vice versa." Although Wilson was confirmed, his remarks raised eyebrows due to widespread skepticism about the alignment of corporate and societal interests. The skepticism of the 1950s looks quaint when compared with today's concerns about whether business leaders will harness the power of artificial intelligence (AI) and workplace automation to pad their own pockets and those of shareholders--not to mention hurting society by causing unemployment, infringing upon privacy, creating safety and security risks, or worse. But is it possible that what is good for society can also be good for business--and vice versa?

It may be the size of a paperclip, but this tiny brain implant has brought life back to men suffering with upper limb paralysis. Australia-based Synchron, a neurovascular bioelectronics medicine company, announced its Stentrode brain computer interface (BCI) has allowed patients to carry out tasks on a computer just by using their mind. Using the implant, patients achieved an average click accuracy of 92 percent and 93 percent and typing speeds of 14 and 20 characters per minute - without lifting a finger. The team is using blood vessels as a natural highway to the brain, which are laced with sensors that record activity. These signals are then sent through a telemetry unit to a small computer taped to the patient's chest, which interprets what actions the individual wants to perform on the nearby PC, such as texting, emailing and shopping online.

A new study has shown that Australians are generally unwilling to sign off on wide-spread use of Artificial intelligence (AI), with less than a quarter of those surveyed approving of the growing technology. The study, conducted by the University of Queensland in partnership with KPMG, shows while 42 per cent generally accept it only 16 per cent approve of AI. More than half of Australians know little about AI and many are unaware that it is being used in everyday applications, like social media. "The benefits and promise of AI for society and business are undeniable," said Professor Nicole Gillespie, KPMG Chair in Organisational Trust and Professor of Management at the University of Queensland Business School. "AI helps people make better predictions and informed decisions, it enables innovation, and can deliver productivity gains, improve efficiency, and drive lower costs. Through such measures as AI-driven fraud detection, it is helping protect physical and financial security – and facilitating the current global fight against COVID-19."