The classical approach to AI was to design systems that were rational at run-time: they had explicit representations of beliefs, goals, and plans and ran inference algorithms, online, to select actions. The rational approach was criticized (by the behaviorists) and modified (by the probabilists) but persisted in some form. Now the overwhelming success of the connectionist approach in so many areas presents evidence that the rational view may no longer have a role to play in AI. I will examine this question from several perspectives, including whether the rationality is present at design-time and/or at run-time, and whether systems with run-time rationality might be useful from the perspectives of computational efficiency, cognitive modeling and safety. I will present some current research focused on understanding the roles of learning in runtime-rational systems with the ultimate aim of constructing general-purpose human-level intelligent robots.

AAAI Invited Talk: The Role of Rationality in Modern AI

Generative AI models such as ChatGPT and Stable Diffusion have greatly accelerated the development and adoption of AI applications. However, one of the key challenges today is AI hallucination – generative models tend to confabulate plausible even authoritative sounding responses that are incorrect and non-factual, leading to downstream harm if not mitigated. Some findings showed that close to 30% of chatbot responses are hallucinations. Why do they hallucinate? And why popular methods such as RLHF and safety fine-tuning cannot completely mitigate this problem? In this talk, I will give an overview of the known causes of hallucination and methods to mitigate them, including in multimodal and multilingual scenarios. I will also talk about the remaining challenges and future directions in AI hallucination research.

AAAI Invited Talk: Machines Make Up Stuff: Why Do Generative Models Hallucinate?

The last years have seen astonishing progress in the capabilities of generative AI techniques, particularly in the areas of language and visual understanding. Key to the success of these models is the availability of very large sets of images and text along with models that are able to digest such large datasets. Unfortunately, we have not been able to replicate this success in the context of robotics, where robots still struggle to perform seemingly simple tasks such as manipulating objects in the real world. A crucial reason for this problem is the lack of data suitable to train powerful, general models for robot decision making and control. In this talk, I will describe our ongoing efforts toward developing the models and generating the kind of data that might enable us to train foundational robot manipulation skills. To generate large amounts of demonstration data, we sample many object rearrangement tasks in physically realistic simulation environments, generate high quality solutions for them, and then train perception-driven manipulation skills that can be used in unknown, real-world environments. We believe that such skills along with generative AI reasoning can provide robots with the capabilities necessary to succeed across a wide range of applications.

IAAI Invited Talk: Toward Foundational Robot Manipulation Skills

Despite meaningful progress over the past few decades, advances in the self-driving industry have plateaued. Traditional approaches are incredibly labor- and capital-intensive, relying on hand-engineered systems and millions of real-world testing miles. Generative AI offers the much-needed new path to unlocking autonomous vehicles at scale and changing the way these systems work. In this lecture, Waabi CEO and Founder Raquel Urtasun will discuss how leveraging generative AI will evolve the way autonomous driving systems are developed and trained, resulting in faster, safer, and more scalable deployment of transformative self-driving technology worldwide.

2024 Robert S. Engelmore Memorial Lecture Award: Accelerating AVs with the Next Generation of Generative AI

The capabilities, autonomy, and efficiency of software and physical agents have increased remarkably in recent times. However, the rapid advancement of these intelligent systems presents significant challenges for humans when they need to interact, coordinate, or collaborate with them. Intelligent advising agents providing guidance to people in such situations can enhance their decision-making and alleviate cognitive load. Alongside identifying valuable advice, these assisting agents must also deliver explanations to convince humans that the provided advice is advantageous and trustworthy. In the presentation, we will highlight the challenges that emerge when humans need to collaborate with drones, autonomous vehicles, and intelligent systems for negotiation and mediation. Additionally, we will introduce several approaches for developing assisting agents aimed at enhancing human performance and satisfaction.

AAAI Invited Talk: Smart Advice: Intelligent Agents Assisting Humans in the Super AI Era 

Charles and Michael are engaged in a decades-long conversation on AI, education, and AI education, exploring where it has been, where it could go, and why it is such a central part of so much that we all do. Occasionally that conversation has found its way into research and educational spaces, including two ongoing online classes in ML and RL taken by about 3300 students each year, a couple of music videos, and a series of educational videos on computing. They believe that this topic is so important and so serious that it must be addressed with a playful frame of mind—no one makes their best decisions and learns effectively when they think the well-being of the world is at stake. In this presentation, Charles and Michael continue their conversation with EAAI with the help of a moderator, answering questions about their approach to AI education and their outlook for the future. They promise to take it just as seriously as they do everything else… and probably there won’t be any singing.

AAAI/EAAI Patrick Henry Winston Outstanding Educator Award: AI Education in the Age of AI

In their first years and with no instruction, children learn one or more natural languages; they develop a host of skills for moving themselves and manipulating objects; they learn the paths connecting places in their homes, the bonds connecting people in their social networks, and the forms and functions of everyday objects; and they deploy concepts of number, geometry, and causality that continue to guide the reasoning of adults, including mathematicians. How children do this is a great, unanswered question, but research on human infants, children, adults, and non-human animals, using diverse methods from the cognitive, brain, and computational sciences, provides some hints. It reveals at least seven domain-specific cognitive systems that emerge and function in infancy: six systems of core knowledge that are shared with other animals and represent places, objects, animate beings, social beings, number and geometry, and a seventh system that likely is unique to humans and underlies language learning during the first year. These automatic, unconscious, fixed systems function throughout life and provide inputs to the malleable, conscious perceptions and thoughts that power our endlessly inventive human minds.

AAAI Invited Talk: How Children Learn

 How could machines learn as efficiently as humans and animals? How could machines learn how the world works and acquire common sense? How could machines learn to reason and plan? Current AI architectures, such as Auto-Regressive Large Language Models fall short. I will propose a modular cognitive architecture that may constitute a path towards answering these questions. The centerpiece of the architecture is a predictive world model that allows the system to predict the consequences of its actions and to plan a sequence of actions that optimize a set of objectives. The objectives include guardrails that guarantee the system’s controllability and safety. The world model employs a Hierarchical Joint Embedding Predictive Architecture (H-JEPA) trained with self-supervised learning. The JEPA learns abstract representations of the percepts that are simultaneously maximally informative and maximally predictable. The corresponding working paper is available here: https://openreview.net/forum?id=BZ5a1r-kVsf

AAAI/IAAI Invited Talk: Objective-Driven AI: Towards Machines that can Learn, Reason, and Plan

For more than 15 years, my team and I have been focused on AI for social impact, deploying end-to-end systems in areas of public health, conservation and public safety. In this talk, I will highlight the results from our deployments for social impact in public health and conservation, as well as required innovations in integrating machine learning and optimization. Within public health, I will present recent results from our work in India with the world’s two largest mobile health programs for maternal and child care that have served millions of beneficiaries. Additionally, I will highlight results from an earlier project on HIV prevention among youth experiencing homelessness in Los Angeles. Turning to conservation, I will highlight efforts for protecting endangered wildlife in national parks around the globe. To address challenges of ML+optimization common to all of these applications, we have advanced the state of the art in decision-focused learning, restless multi-armed bandits, influence maximization in social networks and green security games. In pushing this research agenda, our ultimate goal is to empower local communities and non-profits to directly benefit from advances in AI tools and techniques.

Underline digital video library