Artificial intelligence

AI agent communication refers to how artificial intelligence (AI) agents interact with each other, humans or external systems to exchange information, make decisions and complete tasks.

Federated learning is a decentralized approach to training machine learning (ML) models. Each node across a distributed network trains a global model using its local data, with a central server aggregating node updates to improve the global model.

Learn how Christian Jacobi explains how integral mainframes are to enterprise IT, and how—with new integrations and enhancements—they are redefining their role in modern IT.

Catastrophic forgetting occurs when neural networks forget previously learned tasks after being trained on new data or undergoing fine-tuning for specific tasks.

A new class of prompting techniques, ranging from atom of thoughts (AoT) to chain of draft (CoD), are emerging to increase the efficiency of reasoning in AI models.

IBM Think sits down with Marco Bill-Peter, Red Hat’s Senior Vice President of Customer Experience and Engagement, to discuss open-source AI, innovation in the enterprise, and why IBM and Red Hat are well-positioned to understand this current landscape.

In this tutorial, we will demonstrate how a team of multiple artificial intelligence (AI) agents can collaborate to complete complex tasks and optimize workflows. We built a python application to explain the orchestration of specialized agents working within a multiagent architecture.

AI agents for HR are AI-powered tools designed to automate and enhance various HR team functions including talent acquisition, employee experience, compliance, payroll management, performance management and other day-to-day tasks.

AI agent learning refers to the process by which an artificial intelligence (AI) agent improves its performance over time by interacting with its environment, processing data and optimizing its decision-making.

Tool calling in LLMs is the ability of the LLM to interact with external tools, services, or APIs. This tutorial shows how to configure tool calling with Ollama.

AI is new. The barriers aren’t. See what you can do to make progress.

Prepare for AI as it’s evolving—by bringing the right focus to three areas.

We asked hundreds of global professionals and found what AI-ready companies do well.

AI agents are no longer simply answering questions. They’re beginning to function more like digital coworkers. See how IBM, NVIDIA and others are using AI agents to help enterprises work smarter, not harder.

A new breed of AI threatens to upend centuries of scientific tradition by conducting experiments faster and at scales impossible for humans to match.

AI agent planning refers to the process by which an artificial intelligence (AI) agent determines a sequence of actions to achieve a specific goal.

The convergence of open source AI and hybrid multicloud is the ultimate strategy for businesses looking to maximize their AI investments.

AI TRiSM, or artificial intelligence (AI) trust, risk and security management (AI TRiSM), “ensures AI model governance, trustworthiness, fairness, reliability, robustness, efficacy and data protection."

Agentic reasoning is a component of AI agents that handles decision-making. It allows artificial intelligence agents to conduct tasks autonomously by applying conditional logic or heuristics, relying on perception and memory, enabling it to pursue goals and optimize for the best possible outcome.

AI agent perception refers to an artificial intelligence (AI) agent’s ability to gather, interpret and process data from its environment to make informed decisions.

A ReAct agent is an AI agent that uses the “reasoning and acting” (ReAct) framework to combine an LLM's chain of thought (CoT) reasoning with external tool use.

In this tutorial, you will experiment with Agentic Chunking strategy using [LangChain](https://www.langchain.com/) and the latest IBM® [Granite™](https://www.ibm.com/granite) model now available on [watsonx.ai™](https://www.ibm.com/products/watsonx-ai). The overall goal will be to perform efficient chunking to effectively implement [retrieval augmented generation (RAG)](https://research.ibm.com/blog/retrieval-augmented-generation-RAG).

AI agent memory refers to an artificial intelligence (AI) system’s ability to store and recall past experiences to improve decision-making, perception and overall performance.

Feature selection is the process of selecting the most relevant features of a dataset to use when building and training a machine learning model.

Devs might not be ready to be replaced by LLMs, after all. And while vibe coding may have started as a joke, experts believe it has staying power.

IBM’s Francesca Rossi and a growing number of AI researchers argue that deep learning alone won’t unlock artificial general intelligence (AGI). So what will?

In this episode, Parul Mishra explains the difference between AI agents and AI assistants, and then explores how they can be a game changer for enterprise productivity.

There are 5 main types of AI agents: simple reflex agents, model-based reflex agents, goal-based agents, utility-based agents and learning agents.

AI-driven assistants and intelligent agents can revolutionize the way developers work.

As artificial intelligence increasingly shapes our world, Andrew Barto, one of its founding fathers, warns against hype and fear.