arXiv subjects

Zhou Hao

Publications and source records attributed to Zhou Hao.

3 recordsLinked to original sources

arXiv2024

Exploring Grokking: Experimental and Mechanistic Investigations

The phenomenon of grokking in over-parameterized neural networks has garnered significant interest. It involves the neural network initially memorizing the training set with zero training error and near-random test error. Subsequent prolonged training leads to a sharp transition from no generalization to perfect generalization. Our study comprises extensive experiments and an exploration of the research behind the mechanism of grokking. Through experiments, we gained insights into its behavior concerning the training data fraction, the model, and the optimization. The mechanism of grokking has been a subject of various viewpoints proposed by researchers, and we introduce some of these perspectives.

cs.LG↗

arXiv2020

Imitation Learning for Autonomous Trajectory Learning of Robot Arms in Space

This work adds on to the on-going efforts to provide more autonomy to space robots. Here the concept of programming by demonstration or imitation learning is used for trajectory planning of manipulators mounted on small spacecraft. For greater autonomy in future space missions and minimal human intervention through ground control, a robot arm having 7-Degrees of Freedom (DoF) is envisaged for carrying out multiple tasks like debris removal, on-orbit servicing and assembly. Since actual hardware implementation of microgravity environment is extremely expensive, the demonstration data for trajectory learning is generated using a model predictive controller (MPC) in a physics based simulator. The data is then encoded compactly by Probabilistic Movement Primitives (ProMPs). This offline trajectory learning allows faster reproductions and also avoids any computationally expensive optimizations after deployment in a space environment. It is shown that the probabilistic distribution can be used to generate trajectories to previously unseen situations by conditioning the distribution. The motion of the robot (or manipulator) arm induces reaction forces on the spacecraft hub and hence its attitude changes prompting the Attitude Determination and Control System (ADCS) to take large corrective action that drains energy out of the system. By having a robot arm with redundant DoF helps in finding several possible trajectories from the same start to the same target. This allows the ProMP trajectory generator to sample out the trajectory which is obstacle free as well as having minimal attitudinal disturbances thereby reducing the load on ADCS.

cs.RO↗

arXiv2016

Searching for $τ\rightarrow μγ$ lepton-flavor-violating decay at super charm-tau factory

We investigate the possibility of searching the lepton-flavor-violating (LFV) $τ\rightarrow μγ$ rare decay at the Super Charm-Tau Factory (CTF). By comparing the kinematic distributions of the LFV signal and the standard model (SM) background, we develop an optimized event selection criteria which can significantly reduce the background events. It is concluded that new $2 σ$ upper limit of about $1.9 \times 10^{-9}$ on $Br(τ\rightarrow μγ)$ can be obtained at the CTF, which is beyond the capability of Super-B factory in searching $τ$ lepton rare decay. Within the framework of the scalar leptoquark model, a joint constraint on $λ_1 λ_2$ and $M_{LQ}$ can be derived from the upper bound on $Br(τ\rightarrow μγ)$. With $1000~ fb^{-1}$ data expected at the CTF, we get $λ_1λ_2 < 7.2 \times 10^{-2}~(M_{LQ} = 800~ {\rm GeV})$ and $M_{LQ} > 900~{\rm GeV}~(λ_1 λ_2 = 9 \times 10^{-2})$ at $95\%$ confidence level (C.L.).

hep-ph↗