Safe reinforcement learning (RL) trains a policy to maximize the task reward while satisfying safety constraints. While prior works focus on the performance optimality, we find that the optimal solutions of many safe RL problems are not robust and safe against carefully designed observational perturbations. We formally analyze the unique properties of designing effective observational adversarial attackers in the safe RL setting. We show that baseline adversarial attack techniques for standard RL tasks are not always effective for safe RL and propose two new approaches - one maximizes the cost and the other maximizes the reward. One interesting and counter-intuitive finding is that the maximum reward attack is strong, as it can both induce unsafe behaviors and make the attack stealthy by maintaining the reward. We further propose a robust training framework for safe RL and evaluate it via comprehensive experiments. This paper provides a pioneer work to investigate the safety and robustness of RL under observational attacks for future safe RL studies. Code is available at: \url{https://github.com/liuzuxin/safe-rl-robustness}

Large language models (LLMs) have shown remarkable reasoning capabilities given chain-of-thought prompts (examples with intermediate reasoning steps). Existing benchmarks measure reasoning ability indirectly, by evaluating accuracy on downstream tasks such as mathematical reasoning. However, it is unclear how these models obtain the answers and whether they rely on simple heuristics rather than the generated chain-of-thought. To enable systematic exploration of the reasoning ability of LLMs, we present a new synthetic question-answering dataset called PrOntoQA, where each example is generated from a synthetic world model represented in first-order logic. This allows us to parse the generated chain-of-thought into symbolic proofs for formal analysis. Our analysis on InstructGPT and GPT-3 shows that LLMs are quite capable of making correct individual deduction steps, and so are generally capable of reasoning, even in fictional contexts. However, they have difficulty with proof planning: When multiple valid deduction steps are available, they are not able to systematically explore the different options.

We live in a technological age--or so we are told. Machines promise to transform every facet of human life: robots will staff factory floors, driverless cars will rule the road, and artificial intelligence will govern weapons systems. Politicians and analysts fret over the consequences of such advances, worrying about the damage that will be done to industries and individuals. Governments, they argue, must help manage the costs of progress. These conversations almost always treat technological change as something to be adapted to, as if it were a force of nature, barreling inexorably into the staid conventions and assumptions of modern life. The pace of change seems irrepressible; new technologies will remake societies. All people can do is figure out how best to cope. Nowhere is this outlook more apparent than in the discussion of automation and its impact on jobs. My local grocery store in rural Utah has hung, with no apparent sense of irony, a sign proclaiming the company's support for U.S. workers above a self-checkout machine, a device that uses technology to replace the labor of an employee with the labor of the customer.

Artificial intelligence judging has become a reality. Last month, a Colombian judge used ChatGPT to generate part of his judicial opinion. Estonia has piloted a robot judge, and the United States. These recent events have sparked a debate about "unethical" uses of A.I. in the judiciary. As the technological hurdles to A.I.-judging recede, the remaining barriers are ones of law and ethics.

TheAppWhisperer platform has been a pivotal cyberspace for mobile artists of all abilities to learn about, to explore, to celebrate and to share mobile artworks. Joanne's compassion, inclusivity, and humility are hallmarks in all that she does, and is particularly evident in the platform she has built. In her words, "We all have the potential to remove ourselves from the centre of any circle and to expand a sphere of compassion outward; to include everyone interested in mobile art, ensuring every artist is within reach", she has said. Promotion of mobile artists and the art form as a primary medium in today's art world, has become her life's focus. She has presented lectures bolstering mobile artists and their art from as far away as the Museum of Art in Seoul, South Korea to closer to her home in the UK at Focus on Imaging.

Building systems that achieve a deeper understanding of language is one of the central goals of natural language processing (NLP). Towards this goal, recent works have begun to train language models on narrative datasets which require extracting the most critical information by integrating across long contexts. However, it is still an open question whether these models are learning a deeper understanding of the text, or if the models are simply learning a heuristic to complete the task. This work investigates this further by turning to the one language processing system that truly understands complex language: the human brain. We show that training language models for deeper narrative understanding results in richer representations that have improved alignment to human brain activity. We further find that the improvements in brain alignment are larger for character names than for other discourse features, which indicates that these models are learning important narrative elements. Taken together, these results suggest that this type of training can indeed lead to deeper language understanding. These findings have consequences both for cognitive neuroscience by revealing some of the significant factors behind brain-NLP alignment, and for NLP by highlighting that understanding of long-range context can be improved beyond language modeling.

Topic modelling with innovative deep learning methods has gained interest for a wide range of applications that includes COVID-19. Topic modelling can provide, psychological, social and cultural insights for understanding human behaviour in extreme events such as the COVID-19 pandemic. In this paper, we use prominent deep learning-based language models for COVID-19 topic modelling taking into account data from emergence (Alpha) to the Omicron variant. We apply topic modeling to review the public behaviour across the first, second and third waves based on Twitter dataset from India. Our results show that the topics extracted for the subsequent waves had certain overlapping themes such as covers governance, vaccination, and pandemic management while novel issues aroused in political, social and economic situation during COVID-19 pandemic. We also found a strong correlation of the major topics qualitatively to news media prevalent at the respective time period. Hence, our framework has the potential to capture major issues arising during different phases of the COVID-19 pandemic which can be extended to other countries and regions.

In this paper, we propose a language-universal adapter learning framework based on a pre-trained model for end-to-end multilingual automatic speech recognition (ASR). For acoustic modeling, the wav2vec 2.0 pre-trained model is fine-tuned by inserting language-specific and language-universal adapters. An online knowledge distillation is then used to enable the language-universal adapters to learn both language-specific and universal features. The linguistic information confusion is also reduced by leveraging language identifiers (LIDs). With LIDs we perform a position-wise modification on the multi-head attention outputs. In the inference procedure, the language-specific adapters are removed while the language-universal adapters are kept activated. The proposed method improves the recognition accuracy and addresses the linear increase of the number of adapters' parameters with the number of languages in common multilingual ASR systems. Experiments on the BABEL dataset confirm the effectiveness of the proposed framework. Compared to the conventional multilingual model, a 3.3% absolute error rate reduction is achieved. The code is available at: https://github.com/shen9712/UniversalAdapterLearning.

Existing Graph Neural Networks (GNNs) compute the message exchange between nodes by either aggregating uniformly (convolving) the features of all the neighboring nodes, or by applying a non-uniform score (attending) to the features. Recent works have shown the strengths and weaknesses of the resulting GNN architectures, respectively, GCNs and GATs. In this work, we aim at exploiting the strengths of both approaches to their full extent. To this end, we first introduce the graph convolutional attention layer (CAT), which relies on convolutions to compute the attention scores. Unfortunately, as in the case of GCNs and GATs, we show that there exists no clear winner between the three (neither theoretically nor in practice) as their performance directly depends on the nature of the data (i.e., of the graph and features). This result brings us to the main contribution of our work, the learnable graph convolutional attention network (L-CAT): a GNN architecture that automatically interpolates between GCN, GAT and CAT in each layer, by adding only two scalar parameters. Our results demonstrate that L-CAT is able to efficiently combine different GNN layers along the network, outperforming competing methods in a wide range of datasets, and resulting in a more robust model that reduces the need of cross-validating.

Printed Electronics (PE) exhibits on-demand, extremely low-cost hardware due to its additive manufacturing process, enabling machine learning (ML) applications for domains that feature ultra-low cost, conformity, and non-toxicity requirements that silicon-based systems cannot deliver. Nevertheless, large feature sizes in PE prohibit the realization of complex printed ML circuits. In this work, we present, for the first time, an automated printed-aware software/hardware co-design framework that exploits approximate computing principles to enable ultra-resource constrained printed multilayer perceptrons (MLPs). Our evaluation demonstrates that, compared to the state-of-the-art baseline, our circuits feature on average 6x (5.7x) lower area (power) and less than 1% accuracy loss.