Home  |  Members  |  Publications  |  Funding  |  Demos

Projects

Neuro-Symbolic and Automated Planning

Collaborative Assistants Neuro-Symbolic AI Trusted AI

Team: Vishal Pallagani, Bharath Chandra, Kausik Lakkaraju, Biplav Srivastava

Description: Collaborative Assistants, popularly known as chatbots provide an easy interface for users to obtain answers for their queries. At AI4Society, we build collaborative assistants for various applications such as information retrieval, answer election based questions, help learn puzzle solving through a series of conversations, and obtain information regarding sensor data.

More Details

Planning Using Fast and Slow AI Architecture Neuro-Symbolic AI Automated Planning

Team: University of South Carolina, IBM Research, University of Udine, University of Brescia

Description: The Fast and Slow AI (SOFAI) architecture draws inspiration from cognitive theories discussed by Daniel Kahneman in ‘Thinking Fast and Slow.’ This project aims to develop AI-supported machines that can replicate human decision-making behaviors and provide assistance through nudging and explanations. By creating a cognitive architecture that encompasses both fast (S1) and slow (S2) decision-making modalities, the team seeks to enhance problem-solving capabilities, outperforming traditional symbolic planners like FastDownward.

More Details

Rubik’s Cube Automated Planning

Team: Bharath Muppasani, Vishal Pallagani, Kausik Lakkaraju, Biplav Srivastava, Forest Agostinelli

Description: Rubik’s Cube (RC) is a popular puzzle that is also computationally hard to solve. In this demonstration, we introduce the first PDDL formulation for the 3-dimension RC and solve it with off-the-shelf Fast-Downward planner. We also create a plan executor and visualizer to show how the plan achieves the intended goal. Our system has two audiences:(a) planning researchers who can explore a hard problem, and (b) RC learners wanting to learn how to solve the puzzle at their own pace.

More Details

Trusted AI

AI Teaming and Fairness Recommender Systems Trusted AI

Team: Siva Likitha Valluru, Sai Teja Paladi, Tarmo Koppel, Biplav Srivastava

Description: We present an AI-based system to aid in team formation for researchers responding to funding agency proposals. Our approach leverages NLP techniques to extract and normalize technical skills from various data sources, facilitating matching and teaming based on constraints. We have gathered initial feedback from university researchers to deploy the prototype system and published a dataset for broader use.

More Details

LLM Testing Trusted AI

Team: Kausik Lakkaraju, Sara Rae Jones, Sai Krishna Revanth Vuruma, Vishal Pallagani, Bharath Muppasani, Biplav Srivastava

Description: Increasingly powerful Large Language Model (LLM) based chatbots, like ChatGPT and Bard, are becoming available to users that have the potential to revolutionize the quality of decision-making achieved by the public. We find that although the outputs of the chatbots are fluent and plausible, there are still critical gaps in providing accurate and reliable information using LLM-based chatbots.

More Details

Rating of AI Systems Neuro-Symbolic AI Trusted AI

Team: Kausik Lakkaraju, Biplav Srivastava, Marco Valtorta

Description: AI systems, such as facial recognition and sentiment analyzers, often show model uncertainty, leading to algorithmic bias, particularly concerning protected attributes like gender and race. This research seeks to understand and reduce bias by establishing causal relationships. Assigning ratings to AI systems based on this helps users make informed choices.

More Details

Applications and Others

Insights with Power Data Time-Series Data Analysis

Team: Bharath Muppasani, Cheyyur Jaya Anand, Chinmayi Appajigowda, Lokesh Johri, Biplav Srivastava

Description: The project aims to identify power usage patterns of any system, like buildings or factories, of interest using the harmonics data obtained from MiDAS IoT sensor. We also make power usage dataset (electricity consumption data and harmonics data) available from 8 institutions in manufacturing, education and medical institutions from the US and India.

More Details


GitHub |  Address