Learning agile soccer skills for a bipedal robot with deep reinforcement learning.

We investigated whether deep reinforcement learning (deep RL) is able to synthesize sophisticated and safe movement skills for a low-cost, miniature humanoid robot that can be composed into complex behavioral strategies. We used deep RL to train a humanoid robot to play a simplified one-versus-one soccer game. The resulting agent exhibits robust and dynamic movement skills, such as rapid fall recovery, walking, turning, and kicking, and it transitions between them in a smooth and efficient manner. It also learned to anticipate ball movements and block opponent shots. The agent's tactical behavior adapts to specific game contexts in a way that would be impractical to manually design. Our agent was trained in simulation and transferred to real robots zero-shot. A combination of sufficiently high-frequency control, targeted dynamics randomization, and perturbations during training enabled good-quality transfer. In experiments, the agent walked 181% faster, turned 302% faster, took 63% less time to get up, and kicked a ball 34% faster than a scripted baseline.

A systematic review and meta-analysis of the relationship between youth drinking, self-posting of alcohol use and other social media engagement (2012-21).

BACKGROUND AND AIM: Social networking sites (SNS) are interactive internet-based social platforms that facilitate information sharing. A growing body of literature on exposure to, and self-posting of, alcohol-related content on SNS has examined the relationship between SNS use and alcohol consumption in young people. This study aims to synthesise the literature exploring the relationship between exposure (i.e. viewing or listening of alcohol-related media) and self-posting (i.e. uploading images or text of alcohol content) of alcohol-related media on SNS on alcohol consumption. METHODS: A pre-registered systematic review was conducted in June 2022 within PubMed, Scopus, PsycINFO and Web of Science. Original prospective and cross-sectional studies assessing youth and young adults (≤ 24 years of age) that measured exposure to alcohol-related media or posting of alcohol-related content on SNS and self-reported alcohol consumption outcomes were included. Meta-analyses were conducted on comparable methodologies. RESULTS: Thirty studies were included (n = 19,386). Meta-analyses of cross-sectional studies showed both greater exposure (five studies; pooled ß = 0.34, 95% confidence interval [CI] = 0.23, 0.44, i2 = 27.7%) and self-posting of alcohol-related content (six studies; pooled ß = 0.57, 95%CI = 0.25,0.88, i2 = 97.8%) was associated with greater alcohol consumption. Meta-analyses of three prospective studies also identified that greater exposure predicted greater future alcohol consumption (three studies; pooled ß = 0.13, 95%CI = 0.11,0.15, i2 = 0.0%). Narrative analyses of studies that could not be meta-analysed due to incompatible methodologies were also conducted. Most studies (all four prospective, one of two cross-sectional) identified positive associations between exposure to alcohol-related content and greater average consumption. Most studies (three of four prospective, four of six cross-sectional) reported a positive association between of alcohol-related self-posting and greater average alcohol consumption. CONCLUSIONS: Both exposure to, and self-posting of, alcohol-related content on social networking sites are positively associated with current average consumption, problem drinking, and drinking frequency.

Viral Vaping: A systematic review and meta analysis of e-cigarette and Tobacco-Related social media content and its influence on youth behaviours and attitudes.

AIMS: E-cigarette and tobacco-related content on social media continues to rise from lax restrictions on both personal and promotional posts. This content has been linked to various mechanisms of increased e-cigarette and tobacco use (i.e., lower risk perceptions and increased susceptibility). This study aimed to synthesis the association between exposure to e-cigarette and tobacco-related content and youth behaviours and attitudes. METHODS: A comprehensive search was conducted on PubMed, Scopus, PsycINFO and Web of Science. Studies published post-2004 reporting effect estimates for exposure or engagement with e-cigarette or tobacco content on social media and behaviour or attitude outcomes were included. RESULTS: Thirty-two studies (N = 274,283, aged 9 to 25 years) were included for synthesis. Meta-analyses revealed significant associations between engagement with tobacco content and use (OR 2.21; 95% CI = 1.27-3.82, p =.005; I2 = 96.4%), exposure to tobacco content and never users' lower risk perceptions (OR 0.68; 95% CI = 0.49-0.91; p =.011; I2 = 78.2%), and exposure to e-cigarette content and use (OR 1.37; 95% CI = 0.99-1.88; p = 0.058; I2 = 64.4%). There was no observed relationship between exposure to tobacco content and ever users' risk perceptions (OR 0.83; 95% CI = 0.61-1.13; p =.231; I2 = 83.5%). Qualitative synthesis found significant associations between tobacco exposure and increased current use and pro-tobacco attitudes; e-cigarette exposure and increased susceptibility and lower risk perceptions; tobacco engagement and increased susceptibility; e-cigarette engagement and increased use; dual exposure and increased susceptibility; and dual engagement and increased dual use. Mixed findings were identified for the influence of e-cigarette exposure on attitudes, tobacco exposure on susceptibility, dual exposure on dual use behaviours, and dual engagement on dual susceptibility. CONCLUSIONS: Findings suggest an association between exposure and engagement to e-cigarette or tobacco products on social media and use or pro-use attitudes among youth. Further substantive research in the area of youth-specific use and attitudes following exposure and engagement with e-cigarette and tobacco content is needed to quantify this association.

Looking back and moving forward: dimensions of coping flexibility divergently predict long-term bereavement outcomes.

BACKGROUND AND OBJECTIVES: Bereavement is a serious public health concern. Some people suffer prolonged and debilitating functional impairment after the death of a loved one. Evidence suggests that flexibility in coping approaches predicts resilience after stressful life events, but its long-term effects after the unique experience of bereavement are unknown. Which strategies of coping flexibility predict better-or worse-adjustment over time for bereaved people and at what times? DESIGN AND METHODS: The present study used path analyses to investigate longitudinal effects of forward-focus and loss-focus coping strategies on symptoms of persistent complex bereavement disorder (PCBD), depression, and posttraumatic stress disorder in a spousally bereaved adult sample (N = 248) at three time-points after the loss (∼3 months, ∼14 months, and ∼25 months). RESULTS: Forward-focus coping demonstrated adaptive utility overall, with sooner effects on PCBD than on depression. By contrast, loss-focus coping demonstrated a delayed-onset, maladaptive pattern. CONCLUSIONS: The findings contribute to the coping flexibility literature by suggesting that the adaptiveness or maladaptiveness of different coping strategies may depend on the context that requires coping. In particular, forward-focus coping may be substantially more advantageous than loss-focus coping in the context of bereavement. Implications, limitations, and future research directions are discussed.

#TurntTrending: a systematic review of substance use portrayals on social media platforms.

AIMS: There is a growing body of literature exploring the types of substance-related content and their portrayals on various social media platforms. We aimed to summarize how content related to substances is portrayed on various social media platforms. METHODS: This systematic review was pre-registered on PROSPERO (ref: CRD42021291853). A comprehensive search was conducted in the databases of PubMed, Scopus, PsycINFO and Web of Science in April 2021. Original qualitative studies published post-2004 that included thematic and sentiment analyses of social media content on tobacco, alcohol, psychostimulant, e-cigarette, cannabis, opiate, stimulant/amphetamine, inhalant and novel psychoactive substance were included. Social media platforms were defined as online web- or application-based platforms that allowed users to generate content and interact via 'liking', comment or messaging features. Only studies that included summative and/or thematic content analyses of substance-related social media content were included. RESULTS: A total of 73 studies, which covered 15 905 182 substance-related posts on Twitter, YouTube, Instagram, Pinterest, TikTok and Weibo, were identified. A total of 76.3% of all substance-related content was positive in its depiction of substance use, with 20.2% of content depicting use negatively. Sentiment regarding opiate use however was commonly negative (55.5%). Most studies identified themes relating to Health, Safety and Harms (65.0%) of substance use. Themes relating to Promotions/Advertisements (63.3%), Informative content (55.0%) and Use behaviours (43.3%) were also frequently identified. CONCLUSIONS: Substance-related content that promotes engagement with substance use or actively depicts use appears to be widely available on social media. The large public presence of this content may have concerning influences on attitudes, behaviours and risk perceptions relating to substance use, particularly among the most vulnerable and heaviest users of social media-adolescents and young adults.

Trajectories of resilience and dysfunction following potential trauma: A review and statistical evaluation.

Given the rapid proliferation of trajectory-based approaches to study clinical consequences to stress and potentially traumatic events (PTEs), there is a need to evaluate emerging findings. This review examined convergence/divergences across 54 studies in the nature and prevalence of response trajectories, and determined potential sources of bias to improve future research. Of the 67 cases that emerged from the 54 studies, the most consistently observed trajectories following PTEs were resilience (observed in: n = 63 cases), recovery (n = 49), chronic (n = 47), and delayed onset (n = 22). The resilience trajectory was the modal response across studies (average of 65.7% across populations, 95% CI [0.616, 0.698]), followed in prevalence by recovery (20.8% [0.162, 0.258]), chronicity (10.6%, [0.086, 0.127]), and delayed onset (8.9% [0.053, 0.133]). Sources of heterogeneity in estimates primarily resulted from substantive population differences rather than bias, which was observed when prospective data is lacking. Overall, prototypical trajectories have been identified across independent studies in relatively consistent proportions, with resilience being the modal response to adversity. Thus, trajectory models robustly identify clinically relevant patterns of response to potential trauma, and are important for studying determinants, consequences, and modifiers of course following potential trauma.

Toward personalizing treatment for depression: predicting diagnosis and severity.

OBJECTIVE: Depression is a prevalent disorder difficult to diagnose and treat. In particular, depressed patients exhibit largely unpredictable responses to treatment. Toward the goal of personalizing treatment for depression, we develop and evaluate computational models that use electronic health record (EHR) data for predicting the diagnosis and severity of depression, and response to treatment. MATERIALS AND METHODS: We develop regression-based models for predicting depression, its severity, and response to treatment from EHR data, using structured diagnosis and medication codes as well as free-text clinical reports. We used two datasets: 35,000 patients (5000 depressed) from the Palo Alto Medical Foundation and 5651 patients treated for depression from the Group Health Research Institute. RESULTS: Our models are able to predict a future diagnosis of depression up to 12 months in advance (area under the receiver operating characteristic curve (AUC) 0.70-0.80). We can differentiate patients with severe baseline depression from those with minimal or mild baseline depression (AUC 0.72). Baseline depression severity was the strongest predictor of treatment response for medication and psychotherapy. CONCLUSIONS: It is possible to use EHR data to predict a diagnosis of depression up to 12 months in advance and to differentiate between extreme baseline levels of depression. The models use commonly available data on diagnosis, medication, and clinical progress notes, making them easily portable. The ability to automatically determine severity can facilitate assembly of large patient cohorts with similar severity from multiple sites, which may enable elucidation of the moderators of treatment response in the future.