Investment Rating - The report maintains a positive outlook on overseas Chinese assets and the media sector, focusing on cyclical, undervalued, and fundamentally supported targets [1][2]. Core Insights - Recent policies aimed at boosting consumption have been frequent, with a focus on cyclical sectors and undervalued assets with strong fundamentals. The government has prioritized expanding domestic demand as a key task for the year [1][2][15]. - The report highlights the resilience of leading internet platforms and the potential for a revaluation of Chinese assets, despite possible short-term volatility following rapid price increases [2][6]. - The report suggests specific stocks to watch, including Tencent Holdings, Pinduoduo, and Wanda Film, among others, emphasizing the importance of fundamental support and low valuations [2][6]. Summary by Sections Recent Developments - The government has launched initiatives to stimulate consumption, including a special action plan to enhance consumer spending and financial support for families [15]. - The media sector has seen a 2.21% increase in the week of March 10-14, driven by cyclical segments such as cinema and advertising [2][6]. Industry Policies and News - The report notes significant policy updates aimed at enhancing consumer finance and overall market conditions, which are expected to positively impact the media sector [15]. - Recent financing events in the gaming sector indicate ongoing investment interest, with notable funding rounds for startups [17][18]. Weekly Data Tracking - The total box office revenue from March 8-14 was 510 million yuan, reflecting a 28.9% decrease from the previous week, attributed to seasonal market trends [19][20]. - The report tracks the performance of television dramas and variety shows, noting shifts in viewership and market share among leading platforms [26][29]. Company Valuation Updates - The report indicates that the A-share media sector has seen price increases across various segments, with Wanda Film showing the highest growth at 10.70% during the tracking period [33].

Investment Rating - The industry investment rating is "Overweight," indicating that the industry is expected to outperform the overall market [12]. Core Insights - The report emphasizes optimism towards domestic AI and IP derivatives as two main investment themes, highlighting the ongoing innovation in AI applications and the potential for consumer recovery driven by government policies [4][2]. - The report suggests that the domestic large model innovation and AI application explosion is just beginning, contrasting it with the two-year trend observed in the US market [4]. - Key recommendations include companies like Alibaba, Kingsoft, Tencent, Xiaomi, Kuaishou, and Cloud Music in the cloud computing and AI application sectors, while also highlighting the rise of domestic IP derivatives with companies like Pop Mart and Alibaba Pictures [4]. Summary by Sections AI Applications - Significant advancements in AI gaming and hardware are noted, with companies like Anuttacon developing AI-driven games and partnerships forming to provide AI lifestyle solutions for families [4]. - The report identifies key players in AI gaming such as Giant Network and Kaein Network, and in AI hardware, it mentions Kid King and Volcano Engine [4]. Domestic IP Derivatives - The rise of domestic IP derivatives is highlighted as a core investment theme, with a shift from manufacturing to product and user-centric thinking among companies [4]. - The report points out the increasing global influence of domestic IP, citing successful works like "Nezha 2" and "Black Myth: Wukong" [4]. Advertising and Media - The report indicates that companies like Focus Media have a safety margin, with short-term attention on consumer recovery and long-term focus on new advertising demands driven by AI applications [4]. - The report also provides a valuation table for key companies, detailing their market capitalization, revenue, and profit forecasts for 2023 to 2025 [6].

Core Insights - Podcast advertising spending continued to grow in Q4, increasing by 19% year-over-year [21] - The average ad load increased from 7.09% in Q3 to 7.35% in Q4 [23] - 1,584 new brands advertised on podcasts for the first time, a 10.2% increase from Q3 [24] Spending Trends - The top ten advertisers in Q4 spent an estimated $124 million, a 33% increase from Q3 [32] - The footwear industry saw a significant spending increase of 145%, with brands like DSW and TOMS leading [25] - Overall monthly spending in Q4 exceeded the average of Q3 by 14%, with December showing the largest increase [29] Industry Growth - The fastest-growing industries in Q4 included footwear (+145%), oral hygiene (+99%), and jewelry (+75%) [36] - The consumer electronics sector also experienced a notable increase of 49% in spending [36] Advertising Load and Placement - The average ad load for the sampled episodes decreased to 8.22% in Q4 from 8.28% in Q3, but increased by 5.9% year-over-year [66] - The most common ad placement for new brands was mid-roll ads [59] - Sports podcasts attracted about one in six new brands in their media plans [26][84] Brand Engagement - Direct response advertising accounted for 43% of overall market spending, while brand awareness represented 54% [76] - The average monthly spending on the top 500 podcasts was $329,000 in Q4 [50] Genre Insights - Sports remained the most popular genre for new brands, with 13% of new brands advertising in this category [84] - The average ad load for sports podcasts increased from 6.48% in Q3 to 8.26% in Q4 [60]

本周观点 传媒：DeepSeek 开源周启动，GPT-4.5 发布，持续关注 AI 应用落地情况、对内容板块的赋能及并购重组主线。 投资逻辑 DeepSeek 开源周启动，GPT-4.5 发布，持续关注 AI 应用落地情况、对内容板块的赋能及并购重组主线。1）内 容：院线进入淡季，关注 AI 对影视剧的赋能；游戏板块，看好有基本面支撑、积极布局 AI 的标的。2）AI：关 注 AI 应用落地情况，看好"Chat"类产品及基于算法的应用/工具。 行业新闻 风险提示 内容上线及表现不及预期风险；宏观经济运行不及预期风险；AI 技术迭代和应用不及预期风险；政策监管风险。 敬请参阅最后一页特别声明 1 1）AI：DeepSeek 开源 5 个项目，大模型持续迭代，继续看好 25 年 AI 应用加速落地。我们认为，以 DeepSeek 为代表的训练成本较低且开源的大模型推动了模型平权，使垂类模型训练、AI 应用落地的实现更加经济， 也使创意落地的可能性大大增加。2 月 24 日，Pixverse V4 发布，可 5 秒生成 AI 视频；2 月 25 日，阿里 Wan2.1 视频生成大模型宣布在 Apache2.0 许可 ...

Core Viewpoint - The article discusses the evolution and significance of reinforcement learning (RL) in China, highlighting key figures and their contributions to the field, particularly focusing on Wang Jun and his influence on the development of RL research and education in China [2][46]. Group 1: Historical Context and Development - Wang Jun's journey in AI began with information retrieval and recommendation systems, where he achieved significant academic recognition [4][8]. - His transition to reinforcement learning was influenced by his experiences in advertising, where he recognized the parallels between decision-making in advertising and RL principles [12][14]. - The establishment of the RL China community marked a pivotal moment in promoting RL research and education in China, addressing the lack of resources and formal education in the field [49][50]. Group 2: Contributions and Innovations - Wang Jun and his students have made substantial contributions to RL, including the development of SeqGAN and IRGAN, which integrate RL with generative adversarial networks for improved performance in various applications [23][24]. - The introduction of multi-agent systems in RL research has been a significant focus, with applications in complex environments such as advertising and gaming [27][28]. - The establishment of MediaGamma allowed for practical applications of RL in real-time advertising, showcasing the commercial viability of RL algorithms [17][18]. Group 3: Educational Initiatives and Community Building - The formation of RL China has facilitated knowledge sharing and collaboration among researchers and students, significantly enhancing the learning environment for RL in China [49][52]. - The publication of "Hands-On Reinforcement Learning" has provided accessible educational resources, bridging the gap between theory and practice for students [53]. - Wang Jun's mentorship has fostered a new generation of RL researchers, emphasizing the importance of exploration and innovation in academic pursuits [26][43]. Group 4: Future Directions and Challenges - The integration of RL with large models and embodied intelligence represents a promising frontier for future research, aiming to address the challenges of generalization across different tasks and environments [56][62]. - The ongoing exploration of RL applications in real-world scenarios, such as robotics and automated decision-making, highlights the potential for RL to impact various industries significantly [61][62]. - Despite setbacks in some projects, the commitment to advancing RL research and its applications remains strong among Wang Jun and his students, indicating a resilient and forward-looking approach to the field [56][62].

以下文章来源于网易数读 ，作者网易数读 网易数读 . 网易旗下栏目，用数据说话。 "六边形战士"的宿命。 作者 | 逻辑猫 来源 | 网易数读 （ID：datablog163） 在大众印象里，广告业几乎等同于创意工厂。从"收礼只收脑白金"传遍大街小巷，到"你的益达"让口香 糖品牌一炮而红，这个行业仿佛掌握了点石成金的秘诀。 这些光环也吸引了大批学子报考广告学，憧憬着毕业后一个创意卖出百万，走上人生巅峰，然而从入学 第一天起，现实便开始与理想割裂。 四年里填满了熬不完的夜，做不完的小组作业，比不完的"大广赛""学院奖"。但最震惊的是，有人甚至 还没等到毕业，专业就消失了。学广告，已经没有未来了吗？ 广告学，正在被淘汰 我们熟知的经典广告语，都诞生在改革春风吹满地后。 1979 年正月初一，上海电视台播出了一则特别的内容——一家三口选购补酒送给爷爷，观众起初以为 是新闻，直到片尾才发现是一则广告，这就是中国第一条电视广告。 这一年中国只有 10 家广告公司，而到 1999 年时，已经有 6.48 万家广告经营单位。这种爆发式增 长，催生了广告学专业的开设潮 [1]。 1983 年，厦门大学率先设立广告学专业。到 ...

游戏、出版、广告、影视，各显神通。 作者 | 贝壳XY 编辑 | 小白 春节将至，风云君先给大家拜个早年，祝大家蛇年大吉，我们新春看消费系列也迎来收官之作——传 媒。 截至2024年末，申万传媒指数较年初上涨2.78%，区间最大涨幅64.25%。乍一看似乎不错，但在去年 强力政策催化和宏观经济调整下，这样的涨幅显然是不够看的，不仅跑输沪深300、上证指数，而且 在31个行业大类中也位列中下游。 9月下旬起，在政策助力下，传媒板块跟随大盘大涨。11月，又受到海外AI应用陆续爆发的影响，国 内AI营销、AI玩具等赛道一时成为市场焦点，带动传媒板块进一步上涨。年尾又随着热度消退而跌 回起点，仅比年初高出2.78%。 聚焦传媒板块前十大公司，包括分众传媒（002027.SZ）、芒果超媒（300413.SZ）、昆仑万维（3004 18.SZ）、三七互娱（002555.SZ）、凤凰传媒（601928.SH）、恺英网络（002517.SZ）、光线传媒 （300251.SZ）、中南传媒（601098.SH）、万达电影（002739.SZ）、东方明珠（600637.SH），表现 多数优于板块整体。 注：剔除被实施风险警示股票S ...

吴泳铭说 AI 算力远远不能满足需求。 蔚来计划竞购奥迪在比利时的电动车工厂。 这是大众在比利时的唯一一家工厂，主要生产奥迪 Q8 e-tron，有约 3000 名员工，因为需求疲软， 大众集团 7 月时考虑关闭该厂。据媒体报道，蔚来近几周参观了工厂，并启动了相关的报价准备程 序，准备在下周一向大众汽车提交正式报价。蔚来收购奥迪工厂可能是为应对欧盟对中国电动车加 征关税。 喜茶发布内部信，退出低价内卷。 9 月 18 日喜茶向事业合伙人发布内部信，主题为《为用户创造差异化的品牌和产品》，信中称喜 茶将 "不做同质化产品、不做单纯的低价内卷"。喜茶认为当前茶饮行业的同质化竞争是在消耗用 户对茶饮产品和品牌的热情，差异化是破局的关键，并指出 "门店规模并不是茶饮行业的关键"。 吴泳铭接管阿里巴巴集团和阿里云智能集团已经一年，今天首次在云栖大会登台演讲。他说大模型 推理成本下降速度已经远远超过摩尔定律，一年来通义千问 API 的调用价格下降了 97%，阿里云 还会继续降价。同时 CPU 主导的计算体系正在加速向 GPU 主导转移，新增算力市场超过 50% 的 需求由 AI 驱动产生，过去一年阿里云投资了大量 AI ...