Search

Establishing robust policies is essential to counter attacks or disturbances affecting deep reinforcement learning (DRL) agents. Recent studies explore state-adversarial robustness and suggest the potential lack of an optimal robust policy (ORP), posing challenges in setting strict robustness constraints. This work further investigates ORP: At first, we introduce a consistency assumption of policy (CAP) stating that optimal actions in the Markov decision process remain consistent with minor perturbations, supported by empirical and theoretical evidence. Building upon CAP, we crucially prove the existence of a deterministic and stationary ORP that aligns with the Bellman optimal policy. Furthermore, we illustrate the necessity of $L^{\infty}$-norm when minimizing Bellman error to attain ORP. This finding clarifies the vulnerability of prior DRL algorithms that target the Bellman optimal policy with $L^{1}$-norm and motivates us to train a Consistent Adversarial Robust Deep Q-Network (CAR-DQN) by minimizing a surrogate of Bellman Infinity-error. The top-tier performance of CAR-DQN across various benchmarks validates its practical effectiveness and reinforces the soundness of our theoretical analysis.

Weakly supervised temporal action localization aims to detect and localize actions in untrimmed videos with only video-level labels during training. However, without frame-level annotations, it is challenging to achieve localization completeness and relieve background interference. In this paper, we present an Action Unit Memory Network (AUMN) for weakly supervised temporal action localization, which can mitigate the above two challenges by learning an action unit memory bank. In the proposed AUMN, two attention modules are designed to update the memory bank adaptively and learn action units specific classifiers. Furthermore, three effective mechanisms (diversity, homogeneity and sparsity) are designed to guide the updating of the memory network. To the best of our knowledge, this is the first work to explicitly model the action units with a memory network. Extensive experimental results on two standard benchmarks (THUMOS14 and ActivityNet) demonstrate that our AUMN performs favorably against state-of-the-art methods. Specifically, the average mAP of IoU thresholds from 0.1 to 0.5 on the THUMOS14 dataset is significantly improved from 47.0% to 52.1%., Comment: Accepted by CVPR2021

La posada en marxa en d'un nou centre generador de mobilitat com el Campus del Besos comporta canvis importants en els patrons de mobilitat de la zona i un increment de l'impacte ambiental. El treball proposa analitzar com les infraestructures de la mobilitat i l'urbanisme donen resposta a les demandes de mobilitat generades i avaluar l'impacte ambiental d'aquests nous usuaris. Per acabar, es volen oferir propostes de millora que s'ajustin a les necessitats reals tant dels estudiants com dels treballadors d'aquest nou equipament.

The effects of propiconazole and seaweed extract seed treatments on the germination of Kentucky bluegrass (Poa pratensis L.) under osmotic and temperature stresses were determined in the laboratory. As polyethylene glycol (PEG) 8000 concentrations was increased from 50 to 250 g/L, percent germination was reduced. Propiconazole at 0, 0.5, 0.9, and 1.9 g a.i./kg of seed did not affect total germination, however, germination was delayed with propiconazole seed treatment. Plumule and radicle lengths of seedlings at the termination of experiment 21 days were inhibited with increasing PEG 8000 concentrations and propiconazole treatments. Germination of seed treated with either propiconazole at 1.9 g a.i./kg of seed or seaweed extract at 13 g of product /kg of seed was determined under a range of temperatures from 15 to 40 C on a thermo gradient table. The optimum temperature for Kentucky bluegrass germination was at 25 C. Germination was decreased at temperature below 25 C and no germination was obtained at temperatures above 35 C. Seed treated with propiconazole or seaweed extract did not stimulate seed germination at the different temperatures.