Just two weeks ago, facial recognition technology seemed unstoppable. At the beginning of this year, for instance, news reports cast a light on the secretive company Clearview AI, which scraped social media sites for photos to build a database of more than more than 3 billion photos, sold to law enforcement. Then came a sea change: On Monday, in a letter to Congress, IBM announced it would stop the sale of "general purpose" facial recognition software. On Wednesday, Amazon announced a one-year moratorium on police use of its Rekognition technology by law enforcement, inviting Congress to "put in place stronger regulations to govern the ethical use" of the technology. Amazon in its statement said that, "Congress appears ready to take on this challenge," referring to the mounting pressure to make fundamental changes to U.S. law enforcement following the killing of George Floyd by the Minneapolis police, and law enforcement's heavy-handed and violent response to the Black Lives Matter protests.

On Wednesday, in a brief blog post, Amazon made a surprising announcement: that it would implement a one-year moratorium on police use of its facial recognition service, Rekognition. The post did not mention the furious nationwide demand for reform in response to the killings of George Floyd, Breonna Taylor, and too many other Black people. But it did cite developments "in recent days" indicating that Congress seemed prepared to implement "stronger regulations to govern the ethical use of facial recognition technology"--regulations that Amazon claims to be advocating for and ready to help shape in the coming year. But Amazon's sudden commitment to ostensibly transformative reform should be taken with a grain of salt hefty enough to unseat a Confederate monument from its rock-solid base. Americans won't receive the privacy and civil rights protections they need because a company like Amazon decides to give them to us.

Microsoft Corp. said it won't sell facial-recognition technology to U.S. police until there is a national law regulating its use, echoing similar commitments from Amazon.com Inc. and International Business Machines Corp. made this week. The trio of technology companies have called for clearer federal rules around the surveillance technology amid widespread concern about its potential for racial bias. Meanwhile, the popular fantasy card game, "Magic: The Gathering," removed several cards it deemed racist or culturally offensive from its database, including one depicting figures in pointed hoods. The Hasbro-subsidiary behind the game also pledged to review all cards for material deemed inappropriate. The moves are the latest public actions by businesses lining up to show their commitment to racial equality.

Amazon SageMaker is a fully managed service that provides developers and data scientists the ability to quickly build, train, and deploy machine learning (ML) models. Tens of thousands of customers, including Intuit, Voodoo, ADP, Cerner, Dow Jones, and Thomson Reuters, use Amazon SageMaker to remove the heavy lifting from the ML process. With Amazon SageMaker, you can deploy your ML models on hosted endpoints and get inference results in real time. You can easily view the performance metrics for your endpoints in Amazon CloudWatch, enable autoscaling to automatically scale endpoints based on traffic, and update your models in production without losing any availability. In many cases, such as e-commerce applications, offline model evaluation isn't sufficient, and you need to A/B test models in production before making the decision of updating models.

Artificial intelligence can make drug discovery and development faster and less risky according to an industry expert, who said understanding complex diseases at the genetic level is now a possibility. Data has always been key to drug development. Pharmaceutical products are approved based on empirical evidence of safety and efficacy that is generated in preclinical experiments and clinical trials. Data is also core to the discovery process, with developers selecting candidates that have shown through experimentation that they interact with their chosen disease targets. In recent years, the way data is used for drug discovery and development has changed.

Bidding in real-time auctions can be a difficult stochastic control task; especially if underdelivery incurs strong penalties and the market is very uncertain. Most current works and implementations focus on optimally delivering a campaign given a reasonable forecast of the market. Practical implementations have a feedback loop to adjust and be robust to forecasting errors, but no implementation, to the best of our knowledge, uses a model of market risk and actively anticipates market shifts. Solving such stochastic control problems in practice is actually very challenging. This paper proposes an approximate solution based on a Recurrent Neural Network (RNN) architecture that is both effective and practical for implementation in a production environment. The RNN bidder provisions everything it needs to avoid missing its goal. It also deliberately falls short of its goal when buying the missing impressions would cost more than the penalty for not reaching it.

In this paper, we propose SAMBA, a novel framework for safe reinforcement learning that combines aspects from probabilistic modelling, information theory, and statistics. Our method builds upon PILCO to enable active exploration using novel(semi-)metrics for out-of-sample Gaussian process evaluation optimised through a multi-objective problem that supports conditional-value-at-risk constraints. We evaluate our algorithm on a variety of safe dynamical system benchmarks involving both low and high-dimensional state representations. Our results show orders of magnitude reductions in samples and violations compared to state-of-the-art methods. Lastly, we provide intuition as to the effectiveness of the framework by a detailed analysis of our active metrics and safety constraints.

Text-based games are long puzzles or quests, characterized by a sequence of sparse and potentially deceptive rewards. They provide an ideal platform to develop agents that perceive and act upon the world using a combinatorially sized natural language state-action space. Standard Reinforcement Learning agents are poorly equipped to effectively explore such spaces and often struggle to overcome bottlenecks---states that agents are unable to pass through simply because they do not see the right action sequence enough times to be sufficiently reinforced. We introduce Q*BERT, an agent that learns to build a knowledge graph of the world by answering questions, which leads to greater sample efficiency. To overcome bottlenecks, we further introduce MC!Q*BERT an agent that uses an knowledge-graph-based intrinsic motivation to detect bottlenecks and a novel exploration strategy to efficiently learn a chain of policy modules to overcome them. We present an ablation study and results demonstrating how our method outperforms the current state-of-the-art on nine text games, including the popular game, Zork, where, for the first time, a learning agent gets past the bottleneck where the player is eaten by a Grue.

Regularization and transfer learning are two popular techniques to enhance generalization on unseen data, which is a fundamental problem of machine learning. Regularization techniques are versatile, as they are task- and architecture-agnostic, but they do not exploit a large amount of data available. Transfer learning methods learn to transfer knowledge from one domain to another, but may not generalize across tasks and architectures, and may introduce new training cost for adapting to the target task. To bridge the gap between the two, we propose a transferable perturbation, MetaPerturb, which is meta-learned to improve generalization performance on unseen data. MetaPerturb is implemented as a set-based lightweight network that is agnostic to the size and the order of the input, which is shared across the layers. Then, we propose a meta-learning framework, to jointly train the perturbation function over heterogeneous tasks in parallel. As MetaPerturb is a set-function trained over diverse distributions across layers and tasks, it can generalize to heterogeneous tasks and architectures. We validate the efficacy and generality of MetaPerturb trained on a specific source domain and architecture, by applying it to the training of diverse neural architectures on heterogeneous target datasets against various regularizers and fine-tuning. The results show that the networks trained with MetaPerturb significantly outperform the baselines on most of the tasks and architectures, with a negligible increase in the parameter size and no hyperparameters to tune.

Time series forecasting and spatiotemporal kriging are the two most important tasks in spatiotemporal data analysis. Recent research on graph neural networks has made substantial progress in time series forecasting, while little attention is paid to the kriging problem---recovering signals for unsampled locations/sensors. Most existing scalable kriging methods (e.g., matrix/tensor completion) are transductive, and thus full retraining is required when we have a new sensor to interpolate. In this paper, we develop an Inductive Graph Neural Network Kriging (IGNNK) model to recover data for unsampled sensors on a network/graph structure. To generalize the effect of distance and reachability, we generate random subgraphs as samples and reconstruct the corresponding adjacency matrix for each sample. By reconstructing all signals on each sample subgraph, IGNNK can effectively learn the spatial message passing mechanism. Empirical results on several real-world spatiotemporal datasets demonstrate the effectiveness of our model. In addition, we also find that the learned model can be successfully transferred to the same type of kriging tasks on an unseen dataset. Our results show that: 1) GNN is an efficient and effective tool for spatial kriging; 2) inductive GNNs can be trained using dynamic adjacency matrices; and 3) a trained model can be transferred to new graph structures.