Summary: This paper considers a different perspective on differentially private algorithms where the goal is to give the strongest privacy guarantee possible, subject to the a constraint on the acceptable accuracy. This is in contrast to the more common setting where the we wish to achieve the highest possible accuracy for a given minimum level of privacy. The paper introduces the idea of ex-post differential privacy which provides similar guarantees to differential privacy, but for which the bound on privacy loss is allowed to depend on the output of the algorithm. They then present a general method for private accuracy constrained empirical risk minimization which combines ideas similar to the noise reduction techniques of Koufogiannis et al. and the Above Threshold algorithm. At a high level, their algorithm computes ERM models for increasing privacy parameters (using the noise reduction techniques, rather than independent instantiations of a private learning algorithm) and then uses the above threshold algorithm to find the strongest privacy parameter satisfying the accuracy constraint.

The problem of identifying to which of a given set of classes objects belong is ubiquitous, occurring in many research domains and application areas, including medical diagnosis, financial decision making, online commerce, and national security. But such assignments are rarely completely perfect, and classification errors occur. This means it is necessary to compare classification methods and algorithms to decide which is ``best'' for any particular problem. However, just as there are many different classification methods, so there are many different ways of measuring their performance. It is thus vital to choose a measure of performance which matches the aims of the research or application. This paper is a contribution to the growing literature on the relative merits of different performance measures. Its particular focus is the critical importance of matching the properties of the measure to the aims for which the classification is being made.

Machinery for data analysis often requires a numeric representation of the input. Towards that, a common practice is to embed components of structured data into a high-dimensional vector space. We study the embedding of the tuples of a relational database, where existing techniques are often based on optimization tasks over a collection of random walks from the database. The focus of this paper is on the recent FoRWaRD algorithm that is designed for dynamic databases, where walks are sampled by following foreign keys between tuples. Importantly, different walks have different schemas, or "walk schemes", that are derived by listing the relations and attributes along the walk. Also importantly, different walk schemes describe relationships of different natures in the database. We show that by focusing on a few informative walk schemes, we can obtain tuple embedding significantly faster, while retaining the quality. We define the problem of scheme selection for tuple embedding, devise several approaches and strategies for scheme selection, and conduct a thorough empirical study of the performance over a collection of downstream tasks. Our results confirm that with effective strategies for scheme selection, we can obtain high-quality embeddings considerably (e.g., three times) faster, preserve the extensibility to newly inserted tuples, and even achieve an increase in the precision of some tasks.

The problem of selecting the right state-representation in a reinforcement learning problem is considered. Several models (functions mapping past observations to a finite set) of the observations are given, and it is known that for at least one of these models the resulting state dynamics are indeed Markovian. Without knowing neither which of the models is the correct one, nor what are the probabilistic characteristics of the resulting MDP, it is required to obtain as much reward as the optimal policy for the correct model (or for the best of the correct models, if there are several). We propose an algorithm that achieves that, with a regret of order T {2/3} where T is the horizon time.

A machine learning algorithm is the method by which the AI system conducts its task, generally predicting output values from given input data. The two main processes of machine learning algorithms are classification and regression. Machine learning is a branch of artificial intelligence (AI) and computer science which focuses on the use of data and algorithms to imitate the way that humans learn, gradually improving its accuracy. The term "machine learning" was coined by Arthur Samuel, a computer scientist at IBM and a pioneer in AI and computer gaming. Samuel designed a computer program for playing checkers.

You wouldn't conceive of setting up your own SMS messaging stack across 193 countries and god knows how many telecom carriers in a world where Twilio exists. Machine learning (ML) is in a similar scenario; why would you waste time putting together a whole infrastructure unless Machine Learning is key to your program -- which it probably isn't? Slai is claiming to have laid the foundation to a developer-first machine learning platform to address this specific challenge. It gives developers the tools they need to release machine-learning apps swiftly. The company's offering claims to focus on allowing developers to focus on the machine learning models rather than all of the other nonsense that wastes time but doesn't directly add to the application.

Retailers need more decision automation, faster coordination of supply chains, and faster interactions with consumers, which means they will increasingly rely on AI. Automated decisioning systems will soon be making fine-grained micro-decisions on the retailer's behalf, impacting customers, employees, partners, and suppliers. But these systems can't run autonomously -- they need human managers. Every system for making micro-decisions needs to be monitored. Monitoring ensures the decision-making is "good enough" while also creating the data needed to spot problems and systematically improve the decision-making over time.

Digital generated image of data. Lemonade is one of this year's hottest IPOs and a key reason for this is the company's heavy investments in AI (Artificial Intelligence). The company has used this technology to develop bots to handle the purchase of policies and the managing of claims. Then how does a company like this create AI models? Well, as should be no surprise, it is complex and susceptible to failure.