Joint Client Selection and CPU Frequency Control in Wireless Federated Learning Networks with Power Constraints.

Federated learning (FL) represents a distributed machine learning approach that eliminates the necessity of transmitting privacy-sensitive local training samples. However, within wireless FL networks, resource heterogeneity introduces straggler clients, thereby decelerating the learning process. Additionally, the learning process is further slowed due to the non-independent and identically distributed (non-IID) nature of local training samples. Coupled with resource constraints during the learning process, there arises an imperative need for optimizing client selection and resource allocation strategies to mitigate these challenges. While numerous studies have made strides in this regard, few have considered the joint optimization of client selection and computational power (i.e., CPU frequency) for both clients and the edge server during each global iteration. In this paper, we initially define a cost function encompassing learning latency and non-IID characteristics. Subsequently, we pose a joint client selection and CPU frequency control problem that minimizes the time-averaged cost function subject to long-term power constraints. By utilizing Lyapunov optimization theory, the long-term optimization problem is transformed into a sequence of short-term problems. Finally, an algorithm is proposed to determine the optimal client selection decision and corresponding optimal CPU frequency for both the selected clients and the server. Theoretical analysis provides performance guarantees and our simulation results substantiate that our proposed algorithm outperforms comparative algorithms in terms of test accuracy while maintaining low power consumption.

Lupane- and Friedelane-Type Triterpenoids from Celastrus stylosus.

Two new triterpenoids, 30-hydroxylup-20(29)-ene 3ß-caffeate (1) and 24-nor-friedelan-6α,10-dihydroxy-1,2-dioxo-4,7-dien-29-oic acid (2), together with eight known compounds 3-10, were isolated from the roots of Celastrus stylosus. The structures of these compounds were elucidated on the basis of spectroscopic analyses. To the best of our knowledge, this represents the first study on the chemical constituents of C. stylosus. The antiproliferative activities of the triterpenoids against six human cancer cell lines (PANC-1, A549, PC-3, HepG2, SGC-7901, and HCCLM3) were evaluated. Compounds 3, 4, and 10 exhibited comparable activities against PC-3 and HCCLM3 cell lines as the positive control taxol.

Mechanism of formation of sulphur aroma compounds from l-ascorbic acid and l-cysteine during the Maillard reaction.

The sulphur aroma compounds produced from a phosphate-buffered solution (pH 8) of l-cysteine and l-, l-[1-13C] or l-[4-13C] ascorbic acid, heated at 140±2°C for 2h, were examined by headspace SPME in combination with GC-MS. MS data indicated that C-1 of l-ascorbic acid was not involved in the formation of sulphur aroma compounds. The sulphur aroma compounds formed by reaction of l-ascorbic acid with l-cysteine mainly contained thiophenes, thiazoles and sulphur-containing alicyclic compounds. Among these compounds, 1-butanethiol, diethyl disulphide, 5-ethyl-2-methylthiazole, cis and trans-3,5-dimethyl-1,2,4-trithiolane, thieno[2,3-b]thiophene, thieno[3,2-b]thiophene, cis and trans-3,5-diethyl-1,2,4-trithiolane, 1,2,5,6-tetrathiocane, 2-ethylthieno[2,3-b]thiophene, 2,4,6-trimethyl-1,3,5-trithiane and cyclic octaatomic sulphur (S8) were formed solely by l-cysteine degradation, and the rest by reaction of l-ascorbic acid degradation products, such as hydroxybutanedione, butanedione, acetaldehyde, acetol, pyruvaldehyde and formaldehyde with l-cysteine or its degradation products, such as H2S and NH3. A new reaction pathway from l-ascorbic acid via its degradation products was proposed.