This paper proposes a new sequential model learning architecture to solve partially observable Markov decision problems. Rather than compressing sequential information at every timestep as in conventional recurrent neural network-based methods, the proposed architecture generates a latent variable in each data block with a length of multiple timesteps and passes the most relevant information to the next block for policy optimization. The proposed blockwise sequential model is implemented based on self-attention, making the model capable of detailed sequential learning in partial observable settings. The proposed model builds an additional learning network to efficiently implement gradient estimation by using self-normalized importance sampling, which does not require the complex blockwise input data reconstruction in the model learning. Numerical results show that the proposed method significantly outperforms previous methods in various partially observable environments.

The inclusion of intermittent and renewable energy sources has increased the importance of demand forecasting in power systems. Smart meters can play a critical role in demand forecasting due to the measurement granularity they provide. Despite their virtue, smart meters used for forecasting face some constraints as consumers' privacy concerns, reluctance of utilities and vendors to share data with competitors or third parties, and regulatory constraints. This paper examines a collaborative machine learning method, federated learning extended with privacy preserving techniques for short-term demand forecasting using smart meter data as a solution to the previous constraints. The combination of privacy preserving techniques and federated learning enables to ensure consumers' confidentiality concerning both their data, the models generated using it (Differential Privacy), and the communication mean (Secure Aggregation). To evaluate this paper's collaborative secure federated learning setting, we explore current literature to select the baseline for our simulations and evaluation. We simulate and evaluate several scenarios that explore how traditional centralized approaches could be projected in the direction of a decentralized, collaborative and private system. The results obtained over the evaluations provided decent performance and in a privacy setting using differential privacy almost perfect privacy budgets (1.39,$10e^{-5}$) and (2.01,$10e^{-5}$) with a negligible performance compromise.

In November 2001, Terry Pratchett was in Chester, famed for its Roman ruins and well-preserved medieval architecture. Staying at a hotel in the city centre, Pratchett opened the window of his room, and looked across the historic skyline. "I realised I could drop down on to a roof," he wrote later. "And from then on there was a vista of roofs, leads and ledges leading all the way to the end of the street and beyond; there were even little doors and inviting attic windows … There is a line break, and then he adds. "I'm going to have to stop playing this game." Pratchett was not considering a new career as a cat burglar. He was reflecting on his favourite video game – Thief II: The Metal Age. Released in March 2000, Thief II was the sequel to 1998's Thief: The Dark Project, a pioneering stealth game set in a gothic fantasy world. In both games, players donned the cowl of Garrett, a laconic master thief partly inspired by Raymond Chandler's PI Philip Marlowe. Thief charged players with breaking into medieval mansions, rooftop apartments, banks, cathedrals even police stations, stealing as much coin and valuables as they could while avoiding patrols of sword-wielding guards. Pratchett's relationship with video games is well documented. Always technologically savvy, he was an early adopter of PC gaming, and enjoyed everything from Doom to Deus Ex and Call of Duty. He even helped to create a mod (an unofficial add-on) for The Elder Scrolls IV: Oblivion, writing lines of dialogue for a character. But Pratchett held a particular affection for Thief. He played all three games in the series, and often contributed to a Usenet newsgroup named alt.games.thief-dark-project. That newsgroup, analogous to a modern forum, has long since been deactivated, but its posts survive in a Google groups archive. Combined, they provide a fascinating record of Pratchett's evolving relationship with both the Thief series and video games in general. Like so many players who become involved in online communities, he posted because he was stuck. In a post titled: "Help!

GNW The Global Edge Artificial Intelligence Market size was estimated at USD 572.00 million in 2020 and expected to reach USD 701.73 million in 2021, at a CAGR 23.35% to reach USD 2,014.99 million by 2026. Market Statistics: The report provides market sizing and forecast across five major currencies - USD, EUR GBP, JPY, and AUD. It helps organization leaders make better decisions when currency exchange data is readily available. In this report, the years 2018 and 2019 are considered historical years, 2020 as the base year, 2021 as the estimated year, and years from 2022 to 2026 are considered the forecast period. Market Segmentation & Coverage: This research report categorizes the Edge Artificial Intelligence to forecast the revenues and analyze the trends in each of the following sub-markets: Based on Processor, the market was studied across ASIC, CPU, and GPU.

Salesforce has announced the second edition of its Asia Pacific AI Readiness Index, which saw Singapore retain the top spot for AI readiness. Prepared by Access Partnership, the report uses an AI Index to assess the readiness of governments, businesses, and consumers across 11 APAC economies in their adoption of AI technologies. Singapore leads all three indices of AI readiness with an overall score of 65.7, followed by Japan (60.0) and Hong Kong (59.3). The role of AI is increasing, and Salesforce noted it is delivering around 120 billion AI-powered predictions every day, up from 6.5 billion in October 2019. According to PwC, AI will contribute approximately USD15.7 trillion to global GDP by 2030, up from USD2 trillion in 2019.

Over the last years, many 'explainable artificial intelligence' (xAI) approaches have been developed, but these have not always been objectively evaluated. To evaluate the quality of heatmaps generated by various saliency methods, we developed a framework to generate artificial data with synthetic lesions and a known ground truth map. Using this framework, we evaluated two data sets with different backgrounds, Perlin noise and 2D brain MRI slices, and found that the heatmaps vary strongly between saliency methods and backgrounds. We strongly encourage further evaluation of saliency maps and xAI methods using this framework before applying these in clinical or other safety-critical settings.

Recent work [4] analyses the local convergence of Adam in a neighbourhood of an optimal solution for a twice-differentiable function. It is found that the learning rate has to be sufficiently small to ensure local stability of the optimal solution. The above convergence results also hold for AdamW. In this work, we propose a new adaptive optimisation method by extending AdamW in two aspects with the purpose to relax the requirement on small learning rate for local stability, which we refer to as Aida. Firstly, we consider tracking the 2nd moment r_t of the pth power of the gradient-magnitudes. r_t reduces to v_t of AdamW when p=2. Suppose {m_t} is the first moment of AdamW. It is known that the update direction m_{t+1}/(v_{t+1}+epsilon)^0.5 (or m_{t+1}/(v_{t+1}^0.5+epsilon) of AdamW (or Adam) can be decomposed as the sign vector sign(m_{t+1}) multiplied elementwise by a vector of magnitudes |m_{t+1}|/(v_{t+1}+epsilon)^0.5 (or |m_{t+1}|/(v_{t+1}^0.5+epsilon)). Aida is designed to compute the qth power of the magnitude in the form of |m_{t+1}|^q/(r_{t+1}+epsilon)^(q/p) (or |m_{t+1}|^q/((r_{t+1})^(q/p)+epsilon)), which reduces to that of AdamW when (p,q)=(2,1). Suppose the origin 0 is a local optimal solution of a twice-differentiable function. It is found theoretically that when q>1 and p>1 in Aida, the origin 0 is locally stable only when the weight-decay is non-zero. Experiments are conducted for solving ten toy optimisation problems and training Transformer and Swin-Transformer for two deep learning (DL) tasks. The empirical study demonstrates that in a number of scenarios (including the two DL tasks), Aida with particular setups of (p,q) not equal to (2,1) outperforms the setup (p,q)=(2,1) of AdamW.

A critical aspect of human visual perception is the ability to parse visual scenes into individual objects and further into object parts, forming part-whole hierarchies. Such composite structures could induce a rich set of semantic concepts and relations, thus playing an important role in the interpretation and organization of visual signals as well as for the generalization of visual perception and reasoning. However, existing visual reasoning benchmarks mostly focus on objects rather than parts. Visual reasoning based on the full part-whole hierarchy is much more challenging than object-centric reasoning due to finer-grained concepts, richer geometry relations, and more complex physics. Therefore, to better serve for part-based conceptual, relational and physical reasoning, we introduce a new large-scale diagnostic visual reasoning dataset named PTR. PTR contains around 70k RGBD synthetic images with ground truth object and part level annotations regarding semantic instance segmentation, color attributes, spatial and geometric relationships, and certain physical properties such as stability. These images are paired with 700k machine-generated questions covering various types of reasoning types, making them a good testbed for visual reasoning models. We examine several state-of-the-art visual reasoning models on this dataset and observe that they still make many surprising mistakes in situations where humans can easily infer the correct answer. We believe this dataset will open up new opportunities for part-based reasoning.

Machine learning systems deployed in the wild are often trained on a source distribution but deployed on a different target distribution. Unlabeled data can be a powerful point of leverage for mitigating these distribution shifts, as it is frequently much more available than labeled data. However, existing distribution shift benchmarks for unlabeled data do not reflect the breadth of scenarios that arise in real-world applications. In this work, we present the WILDS 2.0 update, which extends 8 of the 10 datasets in the WILDS benchmark of distribution shifts to include curated unlabeled data that would be realistically obtainable in deployment. To maintain consistency, the labeled training, validation, and test sets, as well as the evaluation metrics, are exactly the same as in the original WILDS benchmark. These datasets span a wide range of applications (from histology to wildlife conservation), tasks (classification, regression, and detection), and modalities (photos, satellite images, microscope slides, text, molecular graphs). We systematically benchmark state-of-the-art methods that leverage unlabeled data, including domain-invariant, self-training, and self-supervised methods, and show that their success on WILDS 2.0 is limited. To facilitate method development and evaluation, we provide an open-source package that automates data loading and contains all of the model architectures and methods used in this paper. Code and leaderboards are available at https://wilds.stanford.edu.

In this paper, we approach the problem of semantic search by framing the search task as paraphrase span detection, i.e. given a segment of text as a query phrase, the task is to identify its paraphrase in a given document, the same modelling setup as typically used in extractive question answering. On the Turku Paraphrase Corpus of 100,000 manually extracted Finnish paraphrase pairs including their original document context, we find that our paraphrase span detection model outperforms two strong retrieval baselines (lexical similarity and BERT sentence embeddings) by 31.9pp and 22.4pp respectively in terms of exact match, and by 22.3pp and 12.9pp in terms of token-level F-score. This demonstrates a strong advantage of modelling the task in terms of span retrieval, rather than sentence similarity. Additionally, we introduce a method for creating artificial paraphrase data through back-translation, suitable for languages where manually annotated paraphrase resources for training the span detection model are not available.