Meta Company Interview Questions and Answers

Outliers are data points that significantly differ from the rest of the data, affecting mean, median, and mode.

• Outliers are extreme values in a dataset that do not follow the general trend of the data.

• Outliers can skew the mean towards their value, making it an inaccurate representation of the central tendency.

• Outliers have little to no effect on the median, as it is not influenced by extreme values.

• Outliers can affect the mode by creating a multimodal distribution or shiftin...read more

Freezing of model refers to fixing the weights of certain layers during training to prevent further updates.

• Freezing certain layers means their weights will not be updated during training

• Helps in transfer learning by keeping pre-trained layers intact

• Prevents overfitting on small datasets

• To prevent freezing, only freeze layers that are already trained on similar data

• Use a lower learning rate for frozen layers to prevent drastic changes

RAG Chatbot Model is a single-agent model while Multi-Agentic RAG Model involves multiple agents collaborating.

• RAG Chatbot Model uses a single agent to generate responses to user queries.

• Multi-Agentic RAG Model involves multiple agents working together to provide more diverse and accurate responses.

• In RAG Chatbot Model, the agent is responsible for both retrieving and generating responses.

• In Multi-Agentic RAG Model, different agents specialize in either retrieval or generatio...read more

RAG Chatbot and Multi Agentic RAG are architectures used for building conversational AI systems.

• RAG Chatbot is a retrieval-augmented generation model that combines retrieval-based and generation-based approaches in chatbots.

• Multi Agentic RAG extends RAG Chatbot to support multiple agents collaborating in a conversation.

• These architectures improve the performance and flexibility of conversational AI systems.

• Example: RAG Chatbot can retrieve relevant information from a knowledg...read more