The issue of fairness in recommendation is becoming increasingly essential as Recommender Systems (RS) touch and influence more and more people in their daily lives. In fairness-aware recommendation, most of the existing algorithmic approaches mainly aim at solving a constrained optimization problem by imposing a constraint on the level of fairness while optimizing the main recommendation objective, e.g., click through rate (CTR). While this alleviates the impact of unfair recommendations, the expected return of an approach may significantly compromise the recommendation accuracy due to the inherent trade-off between fairness and utility. This motivates us to deal with these conflicting objectives and explore the optimal trade-off between them in recommendation. One conspicuous approach is to seek a Pareto efficient/optimal solution to guarantee optimal compromises between utility and fairness. Moreover, considering the needs of real-world e-commerce platforms, it would be more desirable if we can generalize the whole Pareto Frontier, so that the decision-makers can specify any preference of one objective over another based on their current business needs. Therefore, in this work, we propose a fairness-aware recommendation framework using multi-objective reinforcement learning (MORL), called MoFIR (pronounced "more fair"), which is able to learn a single parametric representation for optimal recommendation policies over the space of all possible preferences. Specially, we modify traditional Deep Deterministic Policy Gradient (DDPG) by introducing conditioned network (CN) into it, which conditions the networks directly on these preferences and outputs Q-value-vectors. Experiments on several real-world recommendation datasets verify the superiority of our framework on both fairness metrics and recommendation measures when compared with all other baselines. We also extract the approximate Pareto Frontier on real-world datasets generated by MoFIR and compare to state-of-the-art fairness methods.

Objectives The aims of this study were to describe the treatment outcomes following oral administration of a fixed dose (138 MBq; 3.7 mCi) of radioiodine in hyperthyroid cats and to examine the correlation between total thyroxine (TT4) concentrations before and after treatment. Methods This was a retrospective cohort study that documented the TT4 concentration and clinicopathological parameters at the time of diagnosis and after treatment. Logistic regression was used to assess the relationship between TT4 concentrations before and after treatment. The difference in pre- and post-treatment variables between cats that had TT4 concentrations below or within the reference interval (RI) was compared by the Mann–Whitney U-test. Results Of 161 cats, 133 (82.6%) cats had TT4 concentrations within the RI, four (2.5%) cats had TT4 concentrations above the RI and 24 (14.9%) cats had TT4 concentrations below the RI after treatment. The severity of hyperthyroidism at diagnosis, as measured by the percentage of TT4 elevation above the upper limit of the RI, had no impact on the odds of cats having low TT4 concentrations after treatment (odds ratio 1.00; 95% confidence interval 0.96–1.05; P = 0.828). Conclusions and relevance When using an orally administered fixed dose of radioiodine for the treatment of feline hyperthyroidism, TT4 concentrations at diagnosis cannot be used to predict TT4 concentrations after treatment. The proportion of cats with TT4 concentrations below the lower limit of the RI after treatment was 14.9%. Further work is required to optimise oral radioiodine dosing to achieve maximal euthyroid outcomes.

Background Serum symmetric dimethylarginine (SDMA) is a sensitive renal biomarker for detecting early chronic kidney disease (CKD) in nonhyperthyroid cats, but knowledge regarding its performance in hyperthyroid cats remains limited. Objectives To determine the relationship between serum SDMA, creatinine and total thyroxine (TT4) concentrations in hyperthyroid cats before (T0) and 3 months after (T1) receiving a PO fixed dose of radioiodine. Animals Eighty client‐owned hyperthyroid cats. Methods Prospective cohort study. Serum TT4, and SDMA, creatinine concentrations, and urine specific gravity were measured at T0 and T1. Nonparametric tests were used to determine the relationship among SDMA, and creatinine and TT4 concentrations. Agreement between SDMA and creatinine regarding CKD staging at both time points was assessed using Goodman and Kruskal's gamma statistic. Results Mean serum SDMA concentration increased after treatment of hyperthyroidism. However, 21 of 75 cats experienced a decrease in SDMA between T0 and T1, whereas creatinine decreased in only 2 cats. A moderate correlation between SDMA and creatinine was seen at T1 ( r = 0.53; P < .001) but not at T0 ( r = 0.13; P = .25). Where assessable at T1, poor agreement was observed between SDMA and creatinine and CKD stage (Goodman and Kruskal's gamma 0.20; P = .29). Conclusions and clinical importance Discordant outcomes between SDMA and creatinine after radioiodine treatment in cats with hyperthyroidism suggest extrarenal factors may interfere with the reliability of SDMA to adequately reflect renal function. As a result, SDMA should not be interpreted in isolation in hyperthyroid cats treated with radioiodine.

Hyperthyroidism and chronic kidney disease (CKD) are common diseases of geriatric cats, and often occur concurrently. Thus, a thorough understanding of the influence of thyroid function on renal function is of significant value for all feline practitioners. Among other effects, hyperthyroidism causes protein catabolism and increases renal blood flow and glomerular filtration rate (GFR). These effects render traditional renal markers insensitive for the detection of CKD in cats with uncontrolled hyperthyroidism. Furthermore, the development of iatrogenic hypothyroidism with over treatment of hyperthyroidism can be detrimental to renal function and may negatively affect long-term survival. This review discusses important diagnostic considerations of feline hyperthyroidism, as well as key treatment modalities, with an emphasis on the use of radioiodine and the importance of post treatment monitoring of thyroid and renal parameters. In Australia, a common curative treatment for cats with benign hyperthyroidism (i.e. thyroid hyperplasia or adenoma) is a fixed dose of orally administered radioiodine, regardless of the serum total thyroxine concentration at the time of diagnosis. This review discusses the long term outcomes of this standard of care in comparison with current, relevant research literature from around the world. Finally, this review explores the use of symmetric dimethylarginine (SDMA) in assessing renal function before and after treatment in hyperthyroid cats. SDMA correlates well with GFR and creatinine in non-hyperthyroid cats, but our understanding of its performance in hyperthyroid cats remains in its infancy.

No abstract