Glossary

Insights, News & Updates

Encoder-Decoder Architecture

Encoder-Decoder Architecture is a neural network design that transforms one sequence into another. It consists of two main parts: the encoder, which processes and compresses input data into a compact format, and the decoder, which takes that compressed data and produces the output sequence.

Episodic Compression Strategy

Episodic Memory

Episodic Memory in agentic AI refers to a structured memory mechanism that allows autonomous agents to store, retrieve, and reason over discrete past experiences—referred to as episodes.

Ethical AI Design

Ethical AI Design refers to building artificial intelligence (AI) systems that uphold human values, fairness, accountability, transparency, and safety. It involves designing AI to function accurately or efficiently and act responsibly and respectfully toward individuals, society, and the environment.

Experience Encoding Module

Experience Replay (Agents)

Experience Replay is a learning and memory management mechanism in agentic AI systems that allows an autonomous agent to store past experiences and revisit them during training or adaptation cycles.

Explainable AI (XAI)

Explainable AI (XAI) refers to artificial intelligence systems that make their decision-making processes transparent.

Federated Learning

Federated learning is a machine learning technique allowing multiple devices.

Few-shot Chain-of-Thought

Few-shot Chain-of-Thought (CoT) prompting is used in large language models (LLMs). In this strategy, the user provides a few examples of step-by-step reasoning (called chains of thought ) to guide the model toward generating its rationale for new questions.