Showing papers by "Svetha Venkatesh published in 2023"

Memory-Augmented Theory of Mind Network

Abstract: Social reasoning necessitates the capacity of theory of mind (ToM), the ability to contextualise and attribute mental states to others without having access to their internal cognitive structure. Recent machine learning approaches to ToM have demonstrated that we can train the observer to read the past and present behaviours of other agents and infer their beliefs (including false beliefs about things that no longer exist), goals, intentions and future actions. The challenges arise when the behavioural space is complex, demanding skilful space navigation for rapidly changing contexts for an extended period. We tackle the challenges by equipping the observer with novel neural memory mechanisms to encode, and hierarchical attention to selectively retrieve information about others. The memories allow rapid, selective querying of distal related past behaviours of others to deliberatively reason about their current mental state, beliefs and future behaviours. This results in ToMMY, a theory of mind model that learns to reason while making little assumptions about the underlying mental processes. We also construct a new suite of experiments to demonstrate that memories facilitate the learning process and achieve better theory of mind performance, especially for high-demand false-belief tasks that require inferring through multiple steps of changes.

Machine learning-based discovery of vibrationally stable materials

Abstract: Abstract The identification of the ground state phases of a chemical space in the convex hull analysis is a key determinant of the synthesizability of materials. Online material databases have been instrumental in exploring one aspect of the synthesizability of many materials, namely thermodynamic stability. However, the vibrational stability, which is another aspect of synthesizability, of new materials is not known. Applying first principles approaches to calculate the vibrational spectra of materials in online material databases is computationally intractable. Here, a dataset of vibrational stability for ~3100 materials is used to train a machine learning classifier that can accurately distinguish between vibrationally stable and unstable materials. This classifier has the potential to be further developed as an essential filtering tool for online material databases that can inform the material science community of the vibrational stability or instability of the materials queried in convex hulls.

Zero-shot Sim2Real Adaptation Across Environments

Abstract: Simulation based learning often provides a cost-efficient recourse to reinforcement learning applications in robotics. However, simulators are generally incapable of accurately replicating real-world dynamics, and thus bridging the sim2real gap is an important problem in simulation based learning. Current solutions to bridge the sim2real gap involve hybrid simulators that are augmented with neural residual models. Unfortunately, they require a separate residual model for each individual environment configuration (i.e., a fixed setting of environment variables such as mass, friction etc.), and thus are not transferable to new environments quickly. To address this issue, we propose a Reverse Action Transformation (RAT) policy which learns to imitate simulated policies in the real-world. Once learnt from a single environment, RAT can then be deployed on top of a Universal Policy Network to achieve zero-shot adaptation to new environments. We empirically evaluate our approach in a set of continuous control tasks and observe its advantage as a few-shot and zero-shot learner over competing baselines.

BO-Muse: A human expert and AI teaming framework for accelerated experimental design

Abstract: In this paper we introduce BO-Muse, a new approach to human-AI teaming for the optimization of expensive black-box functions. Inspired by the intrinsic difficulty of extracting expert knowledge and distilling it back into AI models and by observations of human behavior in real-world experimental design, our algorithm lets the human expert take the lead in the experimental process. The human expert can use their domain expertise to its full potential, while the AI plays the role of a muse, injecting novelty and searching for areas of weakness to break the human out of over-exploitation induced by cognitive entrenchment. With mild assumptions, we show that our algorithm converges sub-linearly, at a rate faster than the AI or human alone. We validate our algorithm using synthetic data and with human experts performing real-world experiments.