arXiv subjects

Felix Li

Publications and source records attributed to Felix Li.

5 recordsLinked to original sources

arXiv2024

Meta-Designing Quantum Experiments with Language Models

Artificial Intelligence (AI) can solve complex scientific problems beyond human capabilities, but the resulting solutions offer little insight into the underlying physical principles. One prominent example is quantum physics, where computers can discover experiments for the generation of specific quantum states, but it is unclear how finding general design concepts can be automated. Here, we address this challenge by training a transformer-based language model to create human-readable Python code, which solves an entire class of problems in a single pass. This strategy, which we call meta-design, enables scientists to gain a deeper understanding and extrapolate to larger experiments without additional optimization. To demonstrate the effectiveness of our approach, we uncover previously unknown experimental generalizations of important quantum states, e.g. from condensed matter physics. The underlying methodology of meta-design can naturally be extended to fields such as materials science or engineering.

quant-ph↗

arXiv2022

New Parameter Measurements for the Ultra-Short-Period Planet TOI-1807b

The ultra-short-period (USP) planets are exoplanets with very short orbital periods ($\textit{P} < 1$ day), and TOI-1807b is one such planet recently discovered by the TESS mission where it orbits in the TOI-1807 system that is still little known nowadays. In this paper, we re-analyzed the transit light curves of TOI-1807 using the latest TESS data from Sector 49, combined with previous data from Sector 22 and 23. By running the MCMC simulation through all three sectors, we found that our transit model fits the data from Sector 49 the best, and we deduced that TOI-1807b is a Super-Earth with a mass of $2.27^{+0.49}_{-0.58}\, M_\oplus$, a radius of $1.37^{+0.10}_{-0.09}\, R_\oplus$, a density of $0.875^{+0.264}_{-0.285}\, \rho_\oplus$, and a surface temperature of $1499^{+82}_{-129}\, \mathrm{K}$. We confirmed that TOI-1807b orbits at approximately $0.0135^{+0.0013}_{-0.0022}\, \mathrm{AU}$ with a period of $0.54929^{+0.00012}_{-0.00005}\, \mathrm{days}$, which raises the possibility of the planet being tidally locked due to spin-orbit synchronization. In addition, we suggest that TOI-1807b might slowly undergo its orbital decay process, and we further identify that TOI-1807b is in a circular, synchronous orbit and permanently deformed due to tides, leading to $\sim$4% correction in density. Since TOI-1807 is such a young star with an age of only $300 \pm 80\, \mathrm{Myr}$, we also imply that the radiation emitted from active TOI-1807 could be so intense that it might have destroyed most of the atmosphere over the surface of TOI-1807b.

astro-ph.EP↗

arXiv2022

Insights into Pre-training via Simpler Synthetic Tasks

Pre-training produces representations that are effective for a wide range of downstream tasks, but it is still unclear what properties of pre-training are necessary for effective gains. Notably, recent work shows that even pre-training on synthetic tasks can achieve significant gains in downstream tasks. In this work, we perform three experiments that iteratively simplify pre-training and show that the simplifications still retain much of its gains. First, building on prior work, we perform a systematic evaluation of three existing synthetic pre-training methods on six downstream tasks. We find the best synthetic pre-training method, LIME, attains an average of $67\%$ of the benefits of natural pre-training. Second, to our surprise, we find that pre-training on a simple and generic synthetic task defined by the Set function achieves $65\%$ of the benefits, almost matching LIME. Third, we find that $39\%$ of the benefits can be attained by using merely the parameter statistics of synthetic pre-training. We release the source code at https://github.com/felixzli/synthetic_pretraining.

cs.LG↗

arXiv2019

On-Policy Robot Imitation Learning from a Converging Supervisor

Existing on-policy imitation learning algorithms, such as DAgger, assume access to a fixed supervisor. However, there are many settings where the supervisor may evolve during policy learning, such as a human performing a novel task or an improving algorithmic controller. We formalize imitation learning from a "converging supervisor" and provide sublinear static and dynamic regret guarantees against the best policy in hindsight with labels from the converged supervisor, even when labels during learning are only from intermediate supervisors. We then show that this framework is closely connected to a class of reinforcement learning (RL) algorithms known as dual policy iteration (DPI), which alternate between training a reactive learner with imitation learning and a model-based supervisor with data from the learner. Experiments suggest that when this framework is applied with the state-of-the-art deep model-based RL algorithm PETS as an improving supervisor, it outperforms deep RL baselines on continuous control tasks and provides up to an 80-fold speedup in policy evaluation.

cs.LG↗

arXiv2019

Safety Augmented Value Estimation from Demonstrations (SAVED): Safe Deep Model-Based RL for Sparse Cost Robotic Tasks

Reinforcement learning (RL) for robotics is challenging due to the difficulty in hand-engineering a dense cost function, which can lead to unintended behavior, and dynamical uncertainty, which makes exploration and constraint satisfaction challenging. We address these issues with a new model-based reinforcement learning algorithm, Safety Augmented Value Estimation from Demonstrations (SAVED), which uses supervision that only identifies task completion and a modest set of suboptimal demonstrations to constrain exploration and learn efficiently while handling complex constraints. We then compare SAVED with 3 state-of-the-art model-based and model-free RL algorithms on 6 standard simulation benchmarks involving navigation and manipulation and a physical knot-tying task on the da Vinci surgical robot. Results suggest that SAVED outperforms prior methods in terms of success rate, constraint satisfaction, and sample efficiency, making it feasible to safely learn a control policy directly on a real robot in less than an hour. For tasks on the robot, baselines succeed less than 5% of the time while SAVED has a success rate of over 75% in the first 50 training iterations. Code and supplementary material is available at https://tinyurl.com/saved-rl.

cs.LG↗