AI & Machine Learning

Machine learning in science does present problems in academia due to the lack of reproducibility of results. However, scientists are aware of these problems and a push toward more reproducible and interpretable machine learning models is underway. The real breakthrough will be once this has been completed for neural networks. The scientific community must make a concerted effort in order to understand how these algorithms work and how best to use them to ensure reliable, reproducible, and scientifically valid conclusions are made using data-driven methods.

Experts worry that individuals are ceding control over many important decisions to AI-based technologies and tools.  People do so because of the perceived advantages gained by the use of these powerful tools, – e.g., efficiency, convenience, search capabilities.  But, human autonomy and agency may be at risk, sacrificing, to varying degrees, our independence, privacy, and decision-making power.  These concerns will only increase as AI advances continue to permeate just about every aspect of the economy, society and our personal lives.

It is not a one-time decision to choose from cloud vs. on-premise hosting; that’s a question that developers and business administrators can ask themselves at multiple stages during the life cycle of the AI application. There might arise a need to switch from on-premise to cloud or vice-versa. If a business is in the early stages of digital transformation, then the cloud will be the best option to test AI applications with low cost to experiment. And then, by evaluating the requirement of the applications, businesses can adapt, change, or scale the hosting to on-premise if need be.

Machine learning comes in many different flavors. In this post, we will explore supervised and unsupervised learning, the two main categories of machine learning algorithms. Each subset is composed of many different algorithms that are suitable for various tasks. One of the benefits of unsupervised learning is that it doesn’t require the laborious data labeling process that supervised learning must go through. However, the tradeoff is that evaluating the effectiveness of its performance is also very difficult.

ML models interpretability can be seen as the ability to explain or to present in understandable terms to a human. Regardless of the simple definition, technical challenges and the needs of different user communities have made interpretability a subjective and complicated subject. To make it more objective, a taxonomy was adopted that describes models in terms of their complexity, and categorizes interpretability techniques by the global or local scope of explanations they generate, the family of algorithms to which they can be applied, and their ability to promote trust and understanding.

AI made us think again about the ethics and politics of computerized systems. While computerized systems have been around and influential in our lives for half a century at least, their increased use of our data and increased power to make decisions indeed justifies to think again about their ethics and politics. And, actually, it is the ingredients of “data” and “decision” in this newly reborn notion of “algorithm” that explains why there is reason for ethical concerns and political debate, and hence for the call for regulation.

Although a wide range of traditional optimization methods are available for inventory and price management applications, deep reinforcement learning has the potential to substantially improve the optimization capabilities for these and other types of enterprise operations due to impressive recent advances in the development of generic self-learning algorithms for optimal control. In this article, we explore how deep reinforcement learning methods can be applied in several basic supply chain and price management scenarios. This article is structured as a hands-on tutorial that describes how to develop, debug, and evaluate reinforcement learning optimizers using PyTorch and RLlib.

Data quality is critical because data is used for decision making and powering AI models. Models and decisions are only as good as the data behind them, so any lack of confidence in the data means they are less useful in predicting and providing insights, slowing down, and undermining fast decision making. Trust in data is hard to get and easy to lose, so data quality must be maintained for models and dashboards to be useful at all times.

It’s easy to poke fun at others who pondered the automation “crisis” of his era. But as we experience a fresh wave of alarm over the rise of artificial intelligence and other new technologies, it’s important to ask why and how so many clear-eyed, serious people could have been so wrong, what we can learn from their mistakes—and how, in unexpected ways, they were at least partly right. To understand how tomorrow’s technology will change our lives, we need to look at what yesterday’s futurists got wrong—and right.

Data governance encompasses responsibilities and processes for the security and quality of data, which is used by an organization. Policies for data governance have to grow with upcoming technologies, business practices, and emerging laws. Today, companies have to think about how they are going to use data in terms of storage and processing.  The inclusion of AI can change things for the better. With automation, they can enhance the implementation of security and compliance in their data centers. 

Machine learning’s growing adoption in business across industries reflects how effective its algorithms, frameworks and techniques are at solving complex problems quickly. Machine learning and AI-based techniques are the foundation of a broad spectrum of next-generation logistics and supply chain technologies now under development. Learn from this article about the Key takeaways from the series of machine learning market forecasts and market estimates from the last year from different sources.