Election speeches play an integral role in communicating the vision and mission of the candidates. From lofty promises to mud-slinging, the electoral candidate accounts for all. However, there remains an open question about what exactly wins over the voters. In this work, we used state-of-the-art natural language processing methods to study the speeches and sentiments of the Republican candidate, Donald Trump, and Democratic candidate, Joe Biden, fighting for the 2020 US Presidential election. Comparing the racial dichotomy of the United States, we analyze what led to the victory and defeat of the different candidates. We believe this work will inform the election campaigning strategy and provide a basis for communicating to diverse crowds.

Question answering requiring discrete reasoning, e.g., arithmetic computing, comparison, and counting, over knowledge is a challenging task. In this paper, we propose UniRPG, a semantic-parsing-based approach advanced in interpretability and scalability, to perform unified discrete reasoning over heterogeneous knowledge resources, i.e., table and text, as program generation. Concretely, UniRPG consists of a neural programmer and a symbolic program executor, where a program is the composition of a set of pre-defined general atomic and higher-order operations and arguments extracted from table and text. First, the programmer parses a question into a program by generating operations and copying arguments, and then the executor derives answers from table and text based on the program. To alleviate the costly program annotation issue, we design a distant supervision approach for programmer learning, where pseudo programs are automatically constructed without annotated derivations. Extensive experiments on the TAT-QA dataset show that UniRPG achieves tremendous improvements and enhances interpretability and scalability compared with state-of-the-art methods, even without derivation annotation. Moreover, it achieves promising performance on the textual dataset DROP without derivations.

Overparameterized models with millions of parameters have been hugely successful. In this work, we ask: can the need for large models be, at least in part, due to the \emph{computational} limitations of the learner? Additionally, we ask, is this situation exacerbated for \emph{robust} learning? We show that this indeed could be the case. We show learning tasks for which computationally bounded learners need \emph{significantly more} model parameters than what information-theoretic learners need. Furthermore, we show that even more model parameters could be necessary for robust learning. In particular, for computationally bounded learners, we extend the recent result of Bubeck and Sellke [NeurIPS'2021] which shows that robust models might need more parameters, to the computational regime and show that bounded learners could provably need an even larger number of parameters. Then, we address the following related question: can we hope to remedy the situation for robust computationally bounded learning by restricting \emph{adversaries} to also be computationally bounded for sake of obtaining models with fewer parameters? Here again, we show that this could be possible. Specifically, building on the work of Garg, Jha, Mahloujifar, and Mahmoody [ALT'2020], we demonstrate a learning task that can be learned efficiently and robustly against a computationally bounded attacker, while to be robust against an information-theoretic attacker requires the learner to utilize significantly more parameters.

In this work we study the problem of Byzantine-robust learning when data among clients is heterogeneous. We focus on poisoning attacks targeting the convergence of SGD. Although this problem has received great attention; the main Byzantine defenses rely on the IID assumption causing them to fail when data distribution is non-IID even with no attack. We propose the use of Linear Scalarization (LS) as an enhancing method to enable current defenses to circumvent Byzantine attacks in the non-IID setting. The LS method is based on the incorporation of a trade-off vector that penalizes the suspected malicious clients. Empirical analysis corroborates that the proposed LS variants are viable in the IID setting. For mild to strong non-IID data splits, LS is either comparable or outperforming current approaches under state-of-the-art Byzantine attack scenarios. Most real-world applications using learning algorithms are moving towards distributed computation either: (i) Due to some applications being inherently distributed, Federated Learning (FL) for instance, (ii) or to speed up computation and benefit from hardware parallelization.

Experimental design in field robotics is an adaptive human-in-the-loop decision-making process in which an experimenter learns about system performance and limitations through interactions with a robot in the form of constructed experiments. This can be challenging because of system complexity, the need to operate in unstructured environments, and the competing objectives of maximizing information gain while simultaneously minimizing experimental costs. Based on the successes in other domains, we propose the use of a Decision Support System (DSS) to amplify the human's decision-making abilities, overcome their inherent shortcomings, and enable principled decision-making in field experiments. In this work, we propose common terminology and a six-stage taxonomy of DSSs specifically for adaptive experimental design of more informative tests and reduced experimental costs. We construct and present our taxonomy using examples and trends from DSS literature, including works involving artificial intelligence and Intelligent DSSs. Finally, we identify critical technical gaps and opportunities for future research to direct the scientific community in the pursuit of next-generation DSSs for experimental design.

Embedded topic models are able to learn interpretable topics even with large and heavy-tailed vocabularies. However, they generally hold the Euclidean embedding space assumption, leading to a basic limitation in capturing hierarchical relations. To this end, we present a novel framework that introduces hyperbolic embeddings to represent words and topics. With the tree-likeness property of hyperbolic space, the underlying semantic hierarchy among words and topics can be better exploited to mine more interpretable topics. Furthermore, due to the superiority of hyperbolic geometry in representing hierarchical data, tree-structure knowledge can also be naturally injected to guide the learning of a topic hierarchy. Therefore, we further develop a regularization term based on the idea of contrastive learning to inject prior structural knowledge efficiently. Experiments on both topic taxonomy discovery and document representation demonstrate that the proposed framework achieves improved performance against existing embedded topic models.

With the global population expected to exceed 9 billion by 2050, food security is one of the most important objectives of our time. The agricultural economy employs 65–70 per cent of Africa's labour force and typically accounts for 30–40 per cent of GDP according to the World Bank. With the population in Africa estimated to reach about 2.6 billion by 2050, it is now important that agriculture and food systems be reviewed in order to find innovative approaches at improving food production and utilisation to enhance food security. Being a high-priority sector for the African economy, agriculture, broadly comprising farming and forestry, livestock (milk, eggs and meat) and fisheries, is on the verge of massive transformation with a greater focus on technology integration. Considering the spectrum of the sector, agriculture is still mired with challenges spread across the value chain and needs better optimisation of operations.

Aliens, astronauts, time travel--you name it, there's a dazzling sci-fi film about it. That makes compiling a list of the best sci-fi nearly impossible. It's almost impossible to know where to start--or where to stop. To understand where sci-fi films came from, you need to head back to the dawn of the cinema age. Right at the beginning, Metropolis, released in 1927, used groundbreaking visuals to create a reference point for all future urban dystopias--it's no fluke, for example, that the aesthetic of Blade Runner bears more than a passing resemblance to Fritz Lang's prophetic city hellscape. Then along came War of the Worlds (1953), a gripping tale of alien invasion adapted from H. G. Wells' classic novel. In 1964, Dr. Strangelove did more than most films before or since to ossify the fear of a nuclear holocaust. Below is WIRED's ever-evolving selection of the sci-fi movies everyone should watch, from the obscure to the hugely influential. You may also enjoy our guides to the best sci-fi books of all time and the best space movies. This content can also be viewed on the site it originates from. When Alfonso Cuarón wrote the screenplay for Gravity, he wasn't setting out to make a film about space itself. Rather, he was interested in exploring the concepts of adversity and human resilience, with space as a secondary background. But it was hard for audiences to not be wowed by the visuals in this Oscar-winning film about two scientists (George Clooney and Sandra Bullock) who find themselves stranded in space, and what they must endure in order to get safely back to Earth.

CSE graduate student Yu-Wei Chao has been selected to receive a 2016 Google PhD Fellowship to support his work in the area of computer vision and machine learning. This year, Google awarded 39 fellowships to top PhD students in the US and Canada who are doing exceptional work in computer science, related disciplines, or promising research areas. Yu-Wei is a third year PhD student working with Prof. Jia Deng. His research focuses on computer vision and machine learning. He was awarded the Google PhD Fellowship based on his recent work on large-scale visual recognition of human actions.

Google on Tuesday announced a broad swath of updates to its cloud offerings, aiming to capitalize on its strength in artificial intelligence to gain market share from rivals. The new services--announced at Google's Next '22 event--include Vertex AI Vision, which is designed to make it easier to use AI technology such as image recognition. There's also an AI-based service called Translation Hub that translates documents in 135 languages, the Alphabet Inc.-owned company said. Google is beefing up its cloud infrastructure as well, relying on a fourth-generation version of Intel Corp.'s Xeon Scalable processor and Google's custom Intel chip. The company unveiled a new C3 machine series that's powered by the chips, as well as an updated Tensor processing unit that helps accelerate AI functions.