Abstract: Safety verification for reinforcement learning can include receiving a policy generated by deep reinforced learning, where the policy is used in acting in an environment having a set of states. Responsive to determining that the policy is a non-deterministic policy, the non-deterministic policy can be decomposed into a set of deterministic policies. Responsive to determining that a state-transition function associated with the set of states is unknown, the state-transition function can be approximated at least by training a deep neural network and transforming the deep neural network into a polynomial. Using a constraint solver the policy with the state-transition function can be verified. Runtime shielding can be performed.

Abstract: The present invention relates to the technical field of polymer spectral probes, and particularly to a perchloroethylene derivative and use thereof. The perchloroethylene derivative is prepared by reacting a compound A and a perchloroethylene resin. According to the present invention, the fluorescent polymer can be used as a high-selectivity and high-sensitivity enhanced colorimetric and fluorescent polymer probe for Fe3+ and Cr3+ As compared with the organic small molecule spectral probe, the polymer spectral probe has improved mechanical property, film forming property and excellent recyclability, and thus has a strong practicability.

Abstract: The present invention relates to the technical field of polymer spectral probes, and particularly to a perchloroethylene derivative and use thereof. The perchloroethylene derivative is prepared by reacting a compound A and a perchloroethylene resin. According to the present invention, the fluorescent polymer can be used as a high-selectivity and high-sensitivity enhanced colorimetric and fluorescent polymer probe for Fe3+ and Cr3+ As compared with the organic small molecule spectral probe, the polymer spectral probe has improved mechanical property, film forming property and excellent recyclability, and thus has a strong practicability.

Abstract: The present disclosure provides a method, a device, and a system for packet processing, where the method includes: receiving a protocol packet from a downstream device, where the protocol packet carries uplink port information of the downstream device and the user VLAN of the downstream device; learning port types of the present device based on the uplink port information and classifying the user VLAN of the present device and the user VLAN of the downstream device based on the user VLAN of the downstream device; and forwarding the received packet based on configuration information stored on the present device, where the port type or types to which each VLAN type can be added are specified in the configuration information.

Abstract: The present disclosure provides a method, a device, and a system for packet processing, where the method includes: receiving a protocol packet from a downstream device, where the protocol packet carries uplink port information of the downstream device and the user VLAN of the downstream device; learning port types of the present device based on the uplink port information and classifying the user VLAN of the present device and the user VLAN of the downstream device based on the user VLAN of the downstream device; and forwarding the received packet based on configuration information stored on the present device, where the port type or types to which each VLAN type can be added are specified in the configuration information.