The Living Planet Index produced by WWF estimates that wildlife population sizes have dropped by 68 per cent since 1970. The charity advocates the use of artificial intelligence (AI) as a tool of conservation technology to monitor and curb this alarming rate of decline. One of the most useful applications is in acoustic monitoring, recording the sounds of wildlife ecosystems on weatherproof sensors. Many animals, from birds and bats to mammals and even invertebrates, use sound for communication, navigation and territorial defence, providing reams of rich data on how a species population is doing. AI provides a fast and cost-effective way to analyse hours of recordings for patterns of behaviour.

The Pentagon's Joint Artificial Intelligence Center has awarded a $93.3 million contract to General Atomics Aeronautical Systems Inc (GA-ASI), makers of the MQ-9 Reaper, to equip the drone with new AI technology. The aim is for the Reaper to be able to carry out autonomous flight, decide where to direct its battery of sensors, and to recognize objects on the ground. The contract, announced at the end of last month, builds on a successful test earlier this year. In some ways this is not a major development, more of an incremental step using existing technology. What makes it significant is the drone that is being equipped, and what it will be able to do afterwards.

The ubiquity of smart voice assistants has made conversational shopping commonplace. This is especially true for low consideration segments like grocery. A central problem in conversational grocery is the automatic generation of short product titles that can be read out fast during a conversation. Several supervised models have been proposed in the literature that leverage manually labeled datasets and additional product features to generate short titles automatically. However, obtaining large amounts of labeled data is expensive and most grocery item pages are not as feature-rich as other categories. To address this problem we propose a pre-training based solution that makes use of unlabeled data to learn contextual product representations which can then be fine-tuned to obtain better title compression even in a low resource setting. We use a self-attentive BiLSTM encoder network with a time distributed softmax layer for the title compression task. We overcome the vocabulary mismatch problem by using a hybrid embedding layer that combines pre-trained word embeddings with trainable character level convolutions. We pre-train this network as a discriminator on a replaced-token detection task over a large number of unlabeled grocery product titles. Finally, we fine tune this network, without any modifications, with a small labeled dataset for the title compression task. Experiments on Walmart's online grocery catalog show our model achieves performance comparable to state-of-the-art models like BERT and XLNet. When fine tuned on all of the available training data our model attains an F1 score of 0.8558 which lags the best performing model, BERT-Base, by 2.78% and XLNet by 0.28% only, while using 55 times lesser parameters than both. Further, when allowed to fine tune on 5% of the training data only, our model outperforms BERT-Base by 24.3% in F1 score.

Federated learning (FL) allows a server to learn a machine learning (ML) model across multiple decentralized clients that privately store their own training data. In contrast with centralized ML approaches, FL saves computation to the server and does not require the clients to outsource their private data to the server. However, FL is not free of issues. On the one hand, the model updates sent by the clients at each training epoch might leak information on the clients' private data. On the other hand, the model learnt by the server may be subjected to attacks by malicious clients; these security attacks might poison the model or prevent it from converging. In this paper, we first examine security and privacy attacks to FL and critically survey solutions proposed in the literature to mitigate each attack. Afterwards, we discuss the difficulty of simultaneously achieving security and privacy protection. Finally, we sketch ways to tackle this open problem and attain both security and privacy.

Are you prepared for AI implementation? Do you know what your accompanying data strategy should be? If not, it is likely you aren't alone. According to research by Secondmind, 82 per cent of supply chain managers are frustrated by AI systems and tools during the coronavirus pandemic. In its survey of 500-plus supply chain planners and managers across Europe and the United States, 37 per cent cited a lack of reliable data to feed into AI systems as a concern, at a time when accuracy and speed of decision-making were of the essence.

Around the world, artificial intelligence (AI) is automating functions and making new services possible with breakthroughs in low-cost computing power, cloud computing services, growth in big data and advancements in machine learning and related processes. This webinar discussed the current use of AI in low- and middle-income countries (LMICs), along with trends and challenges in business models, barriers to innovation and AI's ethical and responsible use towards achieving the sustainable development goals. This study examines the current use of AI in low- and middle-income countries (LMICs) across Sub-Saharan Africa, North Africa and South and Southeast Asia. The report mapped a sample of 450 start-ups by sector in alignment with the UN Sustainable ...

Generative adversarial network (GAN) is among the most popular deep learning models for learning complex data distributions. However, training a GAN is known to be a challenging task. This is often attributed to the lack of correlation between the training progress and the trajectory of the generator and discriminator losses and the need for the GAN's subjective evaluation. A recently proposed measure inspired by game theory - the duality gap, aims to bridge this gap. However, as we demonstrate, the duality gap's capability remains constrained due to limitations posed by its estimation process. This paper presents a theoretical understanding of this limitation and proposes a more dependable estimation process for the duality gap. At the crux of our approach is the idea that local perturbations can help agents in a zero-sum game escape non-Nash saddle points efficiently. Through exhaustive experimentation across GAN models and datasets, we establish the efficacy of our approach in capturing the GAN training progress with minimal increase to the computational complexity. Further, we show that our estimate, with its ability to identify model convergence/divergence, is a potential performance measure that can be used to tune the hyperparameters of a GAN.

As the scope for robotic applications extends from structured to unstructured and more complex environments, autonomy has become a desideratum for most of today's robots. The practice of handcrafting robots does not give them the capability to cope with unforeseen situations. Although several research contributions have been made towards robot autonomy, we are nowhere near the level of autonomy that is exhibited by animals, even ones at the lowest biological level of organisation. This is because animals are born with innate capabilities, both in their body structure and intelligence, to survive and develop in their milieus; their behaviours and sometimes their morphological traits can evolve to adapt to persistent changes in their habitats. For example, Corcoran et al [1] studied the co-evolutionary battle between the bat and the moth.

Despite the progress made in recent years in addressing natural language understanding (NLU) challenges, the majority of this progress remains to be concentrated on resource-rich languages like English. This work focuses on Persian language, one of the widely spoken languages in the world, and yet there are few NLU datasets available for this rich language. The availability of high-quality evaluation datasets is a necessity for reliable assessment of the progress on different NLU tasks and domains. We introduce ParsiNLU, the first benchmark in Persian language that includes a range of high-level tasks -- Reading Comprehension, Textual Entailment, etc. These datasets are collected in a multitude of ways, often involving manual annotations by native speakers. This results in over 14.5$k$ new instances across 6 distinct NLU tasks. Besides, we present the first results on state-of-the-art monolingual and multi-lingual pre-trained language-models on this benchmark and compare them with human performance, which provides valuable insights into our ability to tackle natural language understanding challenges in Persian. We hope ParsiNLU fosters further research and advances in Persian language understanding.

A crucial requirement for reliable deployment of deep learning models for safety-critical applications is the ability to identify out-of-distribution (OOD) data points, samples which differ from the training data and on which a model might underperform. Previous work has attempted to tackle this problem using uncertainty estimation techniques. However, there is empirical evidence that a large family of these techniques do not detect OOD reliably in classification tasks. This paper puts forward a theoretical explanation for said experimental findings. We prove that such techniques are not able to reliably identify OOD samples in a classification setting, provided the models satisfy weak assumptions about the monotonicity of feature values and resulting class probabilities. This result stems from the interplay between the saturating nature of activation functions like sigmoid or softmax, coupled with the most widely-used uncertainty metrics.