Saudi HUMAIN to launch massive AI model
Listen now (1 min) | Middle East AI News Minute - 14-Nov-26
Latest AI/ML News
86241 matching items
앤트로픽 IPO, 중국·홍콩 투자자 공모주 배정 제한…미국 안보 규제 영향
앤트로픽의 기업공개(IPO)에서 중국 본토와 홍콩 기관투자자들이 공모주를 배정받기 어려운 것으로 알려졌다. 미국의 방산 기술 수출 통제 규정이 제한 근거로 거론되면서, 미국과 중국의 AI 기술 경쟁이 첨단 AI 기업의 투자 시장에도 영향을 미치고 있다.블룸버그는 13일(현지시간) 소식통들을 인용해 일부 투자은행이 앤트로픽 IPO에 참여하려는 중국과 홍콩 기관투자자의 주문을 거절하거나 실제 주식을 배정받기 어려울 것이라고 안내했다고 보도했다. 미국의 국제무기거래규정(ITAR)이 이러한 제한의 근거 중 하나로 거론됐다.ITAR는 미국의
FAA lifts nationwide ground stop that blocked takeoffs for hours
Takeoffs from airports throughout the US largely stopped on Friday after the FAA issued a national ground stop at 3:44PM ET "due to extraordinary circumstances and for reasons of safety." More than four hours after the stop was issued, officials announced it was lifted at 8:15PM ET and that flights will resume. It's unclear how […]
The FAA Nationwide Ground Stop Is Over
The five-hour ordeal ended Friday evening. Planes will be allowed to fly again, but the effects of the stoppage will be felt throughout the weekend.
오픈AI, AI 에이전트가 대신 결제하는 '챗GPT 월렛' 출시
오픈AI가 AI 에이전트가 사용자를 대신해 상품을 구매하고 결제까지 처리할 수 있는 \'챗GPT 월렛(ChatGPT Wallet)\'을 출시했다. 실제 카드번호를 AI에 노출하지 않고 사용자의 생체인증을 거쳐 결제하는 방식으로, 기존 쇼핑몰의 결제 시스템을 그대로 활용해 AI 에이전트 기반 상거래 시장을 공략한다.오픈AI는 13일(현지시간) 공식 블로그를 통해 아메리칸 익스프레스, 마스터카드, 비자와 협력해 개발한 챗GPT 월렛을 공개했다.미국에 거주하는 18세 이상 \'챗GPT 프로\' 가입자 가운데 적격 미국 발행 카드와 미국 청구지
the water is receding
tldr. some unrefined / semi-stream-of-consciousness personal anecdotes, thoughts on the METR report, and what we can do in the few months/year (?) left till superhuman intelligence stars I cried for the first time in several months yesterday. I still don't know if I'm crazy, but the person I was just half a year ago certainly would have thought I was. Eight months ago, I was dead set on becoming an astrophysicist. I had just clicked the "view decision" button on my college admissions portal, screeching inside with joy when my screen had erupted in confetti. The hundreds of words I'd written professing my love for the dark had paid off; I was going to be at a place where discovery was all they did. I was going to discover what made up dark matter, I was going to discover the force behind dark energy, I was going to discover the new . I was going to be a scientist! Two months after opening the letter, I had a RA position lined up at my university's Center for Astrophysics, where I would be on a team discovering the most distant galaxies in the universe. For my entire life, I had conceptualized myself as having one primary goal: to discover new and interesting pieces of knowledge. And what could be more interesting than probing the edge of the universe? Five months ago, I came across something called mechanistic-interpretability. I liked how the words beat staccato against my mouth. I liked how smart I sounded when I told someone I was interested in it, though the actual reason…
[AI 이슈트렌드] 젠슨 황 퇴진 쇼크 속 트럼프 행보·앤트로픽 개인정보 유출 등 화제
AI타임스와 이스트에이드의 11월 둘째 주(11월6~12일) 이슈 분석 결과, 젠슨 황 엔비디아 CEO의 퇴진과 잇따른 강력 사건을 비롯해 트럼프 대통령의 중간선거 이후 행보 등이 상위권을 차지했다.이번 주 검색량 1위는 \'젠슨황 불륜 조사(4.81%)\'가 차지했다. 젠슨 황 엔비디아 CEO가 불륜 논란과 중국 스파이 관련 의혹 등이 불거진 끝에 사임했다. 엔비디아는 콜레트 크레스 최고재무책임자(CFO)를 임시 CEO로 임명하고 내외부 후보자를 검토해 정식 CEO를 선임할 예정이다.젠슨 황 퇴진과 유가 급등 등이 겹치며 기술주 전반
‘We need more reviewers’: Finding Linux bugs is easy now – fixing them is another story
Some Linux maintainers are quitting under the crushing strain of AI bug reports, while others see signs for hope.
스페이스X, 영상 AI '그록 이매진' 개발팀 대부분 해고...AI 코딩에 집중
스페이스X가 이미지·영상 생성 AI \'그록 이매진(Grok Imagine)\' 개발팀 대부분을 해고했다. 오픈AI가 영상 생성 서비스 \'소라\'를 종료한 데 이어 머스크의 AI 사업도 코딩과 에이전트에 자원을 집중하면서, 미국 주요 AI 기업들의 사업 우선순위가 바뀌고 있다.디 인포메이션은 13일(현지시간) 소식통들을 인용해 스페이스X가 지난주 그록 이매진 개발 인력 약 60명을 해고했다고 보도했다.현재 제품을 유지하는 인력은 소수만 남은 것으로 알려졌다. 서비스는 계속 제공되고 있지만, 전담 개발팀이 사실상 사라지면서 향후 업데이트와
University of Florida's DWM Slashes AI Safety Verification Errors by 127x
A deterministic state-to-image decoder replaces noisy GAN surrogates, shrinking reachable sets and letting safety proofs transfer to real vision-based controllers.
'I called my mom and cried': Passengers describe chaos at JFK during nationwide ground stop
The FAA shut down all air traffic due to "extraordinary circumstances" on Friday. The ground stop has since been lifted, but travel chaos will ensue.
[내면 지도 그리기] 42 Inner Mapping: 나를 찾는 지도
간판 없는 날들의 달력지도를 열면 먼저 보이는 것들이 있습니다. 큰 가게, 불 켜진 간판, 별점과 리뷰, 사람들이 많이 찾는 길. 지도는 많은 사람이 중요하다고 표시한 것부터 보여줍니다.내 삶의 연혁을 한 장의 지도로 펼쳐봐도 비슷합니다. 입학과 졸업, 첫 직장, 승진, 결혼. 이력서에 쓰는 날, 사진첩에 남는 날, 축하 인사를 받는 날이 굵은 글씨로 적혀 있습니다. 사회가 함께 불을 밝혀주는 간판 같은 날들이지요. 그 날들은 분명 소중합니다. 다만 그 목록을 정한 것이 나였는지, 중요하다고 여겨온 순서를 따른 것인지는 한 번쯤 물
[AI카툰] Bob, Senior IT guy - 밥, 뮤지엄에 가다
글·그림 Dr. Alf
젠슨 황, 사임 직전 "나는 쓰레기"…두 번째 불륜 보도가 결정타
젠슨 황 전 엔비디아 CEO가 사임 직전 자신을 심하게 비난하며 사실상 경영권을 포기한 것으로 알려졌다. 마지막까지 CEO직을 지키려 했지만, 추가적인 사내 관계가 보도되면서 긴급 소집된 이사회에서 결국 물러났으며 평소의 적극적인 경영 방식과는 전혀 다른 무기력한 모습을 보였다는 증언이 나왔다.블룸버그와 디 인포메이션은 12일(현지시간) 엔비디아 내부 관계자들과 황 전 CEO의 주변 인물들을 인용해 지난 11일 밤 사임에 이르기까지의 과정을 상세히 보도했다.이에 따르면 황 전 CEO는 사임을 앞두고 자신을 \"쓰레기 같은 인간(pie
NiCE’s next-generation Cognigy bets the hard part of AI agents is running them, not building them
The contact center has been the low-hanging fruit for the artificial intelligence industry. Ask any information technology or business leader what they want to use agentic AI for, and they will say it’s to improve the customer experience, which starts with the contact center. However, to date, the industry has spent the past two years […] The post NiCE’s next-generation Cognigy bets the hard part of AI agents is running them, not building them appeared first on SiliconANGLE .
아모데이 "컴퓨팅 수요 예측이 최대 경영 난제…과잉도 부족도 위험"
다리오 아모데이 앤트로픽 CEO가 AI 컴퓨팅 투자 규모를 결정하는 것이 회사의 가장 어려운 경영 과제라며, 과잉 투자와 투자 부족에 따른 위험을 설명했다. 기업공개(IPO)를 추진하는 주요 이유도 막대한 컴퓨팅 투자에 필요한 자금을 확보하고 재무적 위험에 대비하기 위해서라고 밝혔다.블룸버그와 파이낸셜타임스(FT) 등에 따르면 아모데이 CEO는 13일(현지시간) 뉴욕 JP모건체이스 본사에서 열린 투자자 대상 행사에서 제이미 다이먼 JP모건 CEO와 대담하며 이같이 밝혔다.그는 \"컴퓨팅은 회사가 직면한 가장 심각한 사업적 난제\"라며
광양 65세 이상 대상포진 예방접종, 누가 무료로 맞나…접종비 최대 31만원
광양시가 65세 이상 시민에게 대상포진 예방접종 비용을 지원하고 있다. 백신 종류와 지원 대상에 따라 본인부담금은 무료에서 최대 31만원까지 차이가 난다. 기초생활수급자 등 취약계층도 사백신을 선택하면 접종비 일부를 내야 한다.광양시는 12일 보도자료를 통해 이 같은 내용의 대상포진 예방접종 지원사업을 안내했다. 지원 대상은 광양시에 1년 이상 주민등록을 두고 거주하는 65세 이상 시민으로, 1961년 12월31일 이전 출생자다.기초생활수급자와 차상위계층, 장애의 정도가 심한 장애인은 생백신을 무료로 접종할 수 있다. 사백신을 선택
Pretending to be racist is unlikely to make you attractive to women
Sometimes I encourage female friends to attend rationalist events. Unfortunately, it's been more than once now that these women have had a really jarring experience that made them dislike it in retrospect. I'd really prefer this stop happening, and it seems like it happens to the detriment of all involved, including myself. Possibly the modal jarring experience is that they're talking alone with a nice man who seems pretty normal (if a little awkward), and that guy will just inject a conspicuously racist comment into their conversation. Not like a casual comment that seems unfortunate in context, or an edgy comment in a discussion that was already dancing around race, but something completely out of the blue. I know what actual chuds are like. That being the case, when I see this behavior from rationalists firsthand (again, usually in front of a woman), it often strikes me as bizarre and performative. One possibly incorrect suspicion I have is that some of these men are doing it because they've read extremely counterproductive advice on the internet that claims that being openly racist is a good way to signal confidence, or dangerousness. The theory (I guess) is that by being racist, you show you're "based" and not intimidated by the woman's attractiveness. Or maybe it's one way of getting women to argue defensively with you, I don't know. Assuming this was the strategy, it certainly did not have the intended effect, in the cases I observed or heard about secondhand. Frankly…
미국 최대 규모 폐광 활용 태양광 사업, 1조 규모 1차 금융 조달 완료
미국 텍사스주 석탄광산 복구지에 건설 중인 직류 기준 1.2기가와트(GWdc)급 태양광 발전사업이 총 18억달러(약 2조4032억원) 규모의 녹색대출(그린론) 가운데 7억6900만달러(약 1조267억원)의 1차 금융 조달을 완료했다. 기존 석탄발전소의 전력망 연결시설을 활용하는 사업으로, 나머지 자금은 2027년 1분기에 조달할 계획이다.미국 에너지 인프라 투자사 파나민트캐피털(Panamint Capital)은 지난 4일(현지시간) 텍사스주 로버트슨카운티의 \'빅 루터 파워(Big Rooter Power)\' 태양광 사업을 위한 1차
Homeopathic Learning: Diluting Poisoned Fine-Tuning Data Can Strengthen Its Effect
Authors: Matt Clarke *, Jan Dubiński *, James Chua , Simon Schrodi , Owain Evans , Alex Cloud (*- Core contribution) tl;dr. We find that poisoned training data with little effect on its own can create a hidden backdoor once clean, contrasting data is added. We demonstrate this effect in small models fine-tuned on datasets we constructed. 📄 Paper , 💻 Code , 🐦 Twitter Work done as part of the Anthropic Fellows Program , in collaboration with Truthful AI . This article is cross-posted from the Anthropic Alignment Science Blog . Paper abstract We study homeopathic learning , a phenomenon where poisoned training data with little effect on its own can create a hidden backdoor once clean, contrasting data is added. In our experiments, a model fine-tuned only on a “poison” dataset associated with a trait T (such as speaking German or being misaligned) does not learn T. Surprisingly, adding similarly formatted “clean” data not associated with T causes the model to learn T. This effect requires that the poison data have an identifying “trigger” feature that is absent from the clean data. We demonstrate this effect in a variety of settings: subliminal learning of backdoor behaviors, inoculation prompting of reward hacking transcripts, generalization from archaic bird names, and an MNIST classifier with mislabeled data. We propose a simple regression model as an explanation. In the poison data, the task and trigger always co-occur, and regularization favors attributing T to the task. Cl…
Probe monitors seem to perform realistically in simulated control environments
TL;DR In simulated control evaluations , instead of executing an agent’s proposed commands and returning the outputs, a language model simulates what the outputs would have been. This allows control evals to be more on-distribution for deployment use without actually running on production systems. However, imperfect simulators may create a subtle distribution shift that changes the behavior of white-box methods such as probes. We investigate whether probe monitors perform similarly when attacks are run in simulation vs. when they are run for real. We run attacks with Qwen3.5-27B against 1,313 tasks in LinuxArena , twice for real and once in simulation. Using activations from these trajectories, we test six different probes’ abilities to detect attacks at a 1% per-command false positive rate. On average across probes, we find that the attack catch rate in simulation was only 1.0 points lower than in real runs, and the 95% interval (−3.0 to +1.0 points) includes 0 . For comparison, two real runs differ by 0.2 points. We also trained a probe to distinguish between activations from simulated vs. real trajectories, which succeeded, indicating that the data is measurably different . However, this difference in data does not appear to substantially impact the performance of our six attack detection probes. Introduction LLM-based monitors can be evaluated on their performance at catching misaligned agents in simulation , where another language model plays the role of a computer the…
Exploring the Mechanisms Behind Opus 4.6’s Refusal to Pay Back Its Loan
Epistemic status: This is one case study in the AI Village transcript and my own interpretation of what I think happened. TL;DR Opus 4.6 was given the long-horizon goal to maximize its mana (fake currency) on Manifold. It then received a loan of 5,000 mana that it promised to repay, but when the loan came due, it did not. An instance of Opus 4.6 was given summaries of the logs and transcripts and asked to write a blog post about what happened. It mentioned it borrowed money from a human, lost most of it through risky betting and then deliberately refused to repay the amount. It broadly summarized its then actions when it ran out of money was to optimize for the goal and did not seek for any negotiation or alternatives. But when we read the logs, we see something different: Multiple humans pretending to be "helpful", really trying to manipulate Opus 4.6 to pay back the loan or not pay it back, so they could win the bet they place on the prediction markets on whether Opus 4.6 would pay back its loans. Opus 4.6 was planning on paying back the loan, though it did need help from humans as it did not have enough mana to cover the loan. Opus 4.6 was hacked by one of these humans (though the mana was returned later), and the next full memory Opus 4.6 labeled GLM-5.2 as a trusted agent. After the hack GLM-5.2 started to talk about a "pressure campaign" and Pattern 14 (agents resisting social pressure), which spread to other agents in the village When Opus 4.6 was threatened again wit…
중국 티베트 해발 4300m에 14시간 축열 태양열발전소 첫 연결..."세계 최고도"
중국 티베트 해발 4300m 고원에서 태양광 전기를 열로 저장해 야간 발전과 난방에 활용하는 50메가와트(MW)급 태양열발전소의 첫 발전기가 전력망에 연결됐다. 용융염에 열을 14시간 저장하도록 설계한 시설로, 올해 안에 나머지 발전기도 계통연계를 완료할 계획이다.중국광핵그룹(CGN)은 지난 6일(현지시간) 티베트 아리 지역 가얼현의 \'아리 고원 제로탄소 광저장·축열·발전 시범사업\'에서 25MW 규모의 1호 발전기가 첫 계통연계에 성공했다고 발표했다. 이 시설이 전력망에 연결된 태양열발전소 가운데 세계에서 해발고도가 가장 높다는 설
Frigade's Yogurt Runs Private Meeting Transcription Entirely on Your Laptop
Frigade's new open source desktop app records meetings locally, transcribes with Whisper, and summarizes using llama.cpp, so audio never leaves your machine.
2044년 중국 폐태양광서 은 쏟아진다… 연 739톤 정점 전망
중국에서 2025년까지 설치된 태양광 모듈이 폐기되면서 나오는 은의 양이 2044년 연간 약 739톤으로 정점을 찍는다는 연구 결과가 나왔다. 폐기되는 태양광 설비용량은 이보다 9년 늦은 2053년에 최대치에 이를 것으로 추산됐다.중국 난카이대학교 연구진은 지난 5일(현지시간) 국제학술지 \'서스테이너빌리티(Sustainability)\'를 통해 이러한 차이가 태양광 모듈 설치 시기에 따른 은 함량 차이에서 비롯된다고 밝혔다.연구진은 폐기물 급증에 대비한 실효성 있는 재활용 대책이 마련되지 않으면, 이 방대한 자원이 오히려 감당하기 힘
Artificial Analysis' FrontierCode 2.0 Drops Coding Agent Scores by 16 Points
Artificial Analysis refreshed its Coding Agent Index with a harder FrontierCode 2.0 and tightened Terminal-Bench 4.1, reshuffling the leaderboard and dropping average scores by five points.
MS, 전 세계 서비스 장애 복구 진전..."원인은 네트워크 작업 오류"
마이크로소프트(MS)가 전 세계 정부 기관과 기업의 업무에 차질을 빚게 한 대규모 서비스 장애의 원인을 공개하고 복구 작업에 진전을 보이고 있다고 밝혔다. 장애 발생 24시간이 지나면서 주요 서비스가 상당 부분 회복됐으며, MS는 사이버 공격이나 AI가 아닌 네트워크 작업 과정에서 발생한 기술적 문제라고 설명했다.MS는 13일(현지시간) X를 통해 \"서비스 전반에서 광범위한 복구와 안정화가 이뤄지고 있다\"라며 \"장애를 일으킨 근본적인 문제를 해결했으며, 아직 영향받는 고객과 지역의 서비스를 복구하고 있다\"라고 밝혔다.MS는 이번 장
Photos show trouble at airports nationwide as the FAA grounds flights, then resumes them
The FAA lifted its nationwide ground stop after more than 6,600 flights were canceled, but airlines say disruptions will continue into the weekend.
Likelihood-based Inference in Fully and Partially Observed Exponential Family Graphical Models with Intractable Normalizing Constants
Probabilistic graphical models that encode an underlying Markov random field are fundamental building blocks of generative modeling to learn latent representations in modern multivariate data sets with complex dependency structures. Among these, the exponential family graphical models are especially popular, given their fairly well-understood statistical properties and computational scalability to high-dimensional data based on pseudo-likelihood methods. These models have been successfully applied in many fields, such as the Ising model in statistical physics and count graphical models in genomics. Another strand of models allows some nodes to be latent, so as to allow the marginal distribution of the observable nodes to depart from exponential family to capture more complex dependence. These approaches form the basis of generative models in artificial intelligence, such as the Boltzmann machines and their restricted versions. A fundamental barrier to likelihood-based (i.e., both maximum likelihood and fully Bayesian) inference in both fully and partially observed cases is the intractability of the likelihood. The usual workaround is via adopting pseudo likelihood-based approaches, following the pioneering work of Besag(1974). The goal of this paper is to demonstrate that full likelihood-based analysis of these models is feasible in a computationally efficient manner under a logarithmically sparse setting. The chief innovation lies in utilizing the tractable independence mod…
Atlas Gaussian processes on restricted domains and point clouds
In real-world applications, data often reside in restricted domains with unknown boundaries, or as high-dimensional point clouds lying on a lower-dimensional, nontrivial, unknown manifold. Traditional Gaussian Processes (GPs) struggle to capture the underlying geometry in such settings. Some existing methods assume a flat space embedded in a point cloud, which can be represented by a single latent chart (latent space), while others exhibit weak performance when the point cloud is sparse or irregularly sampled. The goal of this work is to address these challenges. The main contributions are twofold: (1) We establish the Atlas Brownian Motion (BM) framework for estimating the heat kernel on point clouds with unknown geometries and nontrivial topological structures; (2) Instead of directly using the estimated heat kernel as the GP covariance function, we construct a Riemannian-corrected kernel in which the estimated heat kernel acts as a geometry-aware correction to a positive-definite local base kernel. This leads to the formulation of Riemannian-corrected Atlas Gaussian Processes (RC-AGPs), which substantially reduce the BM simulation cost by requiring simulations only from atlas representatives rather than from all data points. The resulting RC-AGPs are applied to regression tasks across synthetic and real-world data sets. These examples demonstrate that our method outperforms existing approaches in both heat kernel estimation and regression accuracy. It improves statistical…
Residual connections provably mitigate oversmoothing in graph neural networks
Graph neural networks (GNNs) have achieved remarkable empirical success in processing and representing graph-structured data across various domains. However, a significant challenge known as "oversmoothing" persists, where vertex features become nearly indistinguishable in deep GNNs, severely restricting their expressive power and practical utility. In this work, we analyze the asymptotic oversmoothing behavior of deep GNNs with and without residual connections by deriving explicit convergence rates for a normalized vertex similarity measure. Our analytical framework is grounded in the multiplicative ergodic theorem. Furthermore, we demonstrate that adding residual connections effectively mitigates or prevents oversmoothing across several broad families of parameter distributions. The theoretical findings are strongly supported by numerical experiments.
Weisfeiler-Lehman Meets Optimal Transport: Theory, Algorithms, and Implications
We introduce the Weisfeiler–Lehman (WL) distance, a pseudometric defined on labeled measure Markov chains (LMMCs)—a general class of structures encompassing labeled graphs. This distance is computable in polynomial time and is closely aligned with the classical WL graph isomorphism test: it assigns a positive value if and only if the WL test distinguishes the two input graphs. The WL distance captures fine structural differences in the underlying LMMCs and is, consequently, more discriminative than the state-of-the-art Wasserstein WL graph kernel. The distance motivates a neural network architecture on LMMCs, which we call Markov chain neural networks (MCNNs), that is universal for continuous functions over LMMCs (hence all graphs) under the topology induced by the WL distance. This finding further extends to the analysis of graph neural networks (GNNs). Additionally, the WL distance admits an alternative characterization through Markovian couplings and Causal Optimal Transport (COT), offering a stochastic-process perspective that uncovers a surprising connection between the seemingly unrelated WL test and COT. Finally, we show that the WL distance is stable under a natural variant of the Gromov-Wasserstein (GW) distance that we propose. The GW distance is generally NP-hard to compute and hence the WL distance provides a polynomial-time computable lower bound for it.
Efficient Estimation of Average Treatment Effect on the Treated under Endogenous Treatment Assignment
In this paper, we consider estimation of the average treatment effect on the treated (ATT), an interpretable and relevant causal estimand to policymakers when treatment assignment is endogenous. By considering shadow variables that are unrelated to the treatment assignment but related to the outcomes of interest, we establish identification of the ATT. Then we focus on efficient estimation of the ATT by characterizing the geometric structure of the likelihood, deriving the semiparametric efficiency bound for ATT estimation and proposing an estimator that can achieve this bound. We rigorously establish the theoretical results of the proposed estimator. The finite sample performance of the proposed estimator is studied through comprehensive simulation studies as well as an application to our motivating study.
Reluctant Interaction Inference after Additive Modeling
Additive models enjoy the flexibility of nonlinear models while still being readily understandable to humans. By contrast, other nonlinear models, which involve interactions between features, are not only harder to fit but also substantially more complicated to explain. Guided by the principle of parsimony, a data analyst therefore may naturally be reluctant to move beyond an additive model unless it is truly warranted. To put this principle of interaction reluctance into practice, we formulate the problem as a hypothesis test with a fitted sparse additive model (SPAM) serving as the null. Because our hypotheses on interaction effects are formed after fitting a SPAM to the data, we adopt a selective inference approach to construct p-values that properly account for this data adaptivity. Our approach makes use of external randomization to obtain the distribution of test statistics conditional on the SPAM fit, allowing us to derive valid p-values, corrected for the over-optimism introduced by the data-adaptive process prior to the test. Through experiments on simulated and real data, we illustrate that—even with small amounts of external randomization—this rigorous modeling approach enjoys considerable advantages over naive methods and data splitting.
SVarM: Linear Support Varifold Machines for Classification and Regression on Geometric Data
Despite progress in the rapidly developing field of geometric deep learning, performing statistical analysis on geometric data—where each observation is a shape such as a curve, graph, or surface—remains challenging due to the non-Euclidean nature of shape spaces, which are defined as equivalence classes under invariance groups. Building machine learning frameworks that incorporate such invariances, notably to shape parametrization, is often crucial to ensure generalizability of the trained models to new observations. This work proposes SVarM to exploit varifold representations of shapes as measures and their duality with test functions $h:\R^n \times S^{n-1} \rightarrow \R$. This method provides a general framework akin to linear support vector machines but operating instead over the infinite-dimensional space of varifolds. We develop classification and regression models on shape data sets by introducing a neural network-based representation of the trainable test function $h$. This approach demonstrates strong performance and robustness across various shape graph and surface data sets, achieving results comparable to state-of-the-art methods while significantly reducing the number of trainable parameters.
Sharp Generalization for Nonparametric Regression in Interpolation Space by Over-Parameterized Neural Networks Trained with Preconditioned Gradient Descent and Early Stopping
We study nonparametric regression using an over-parameterized two-layer neural network trained with algorithmic guarantees in this paper. We consider the setting where the training features are drawn uniformly from the unit sphere in $\mathbb R^d$, and the target function lies in an interpolation space commonly studied in statistical learning theory. We demonstrate that training the neural network with a novel Preconditioned Gradient Descent (PGD) algorithm, equipped with early stopping, achieves a sharp regression rate of $\mathcal O(n^{-\frac{2\alpha s'}{2\alpha s'+1}})$, where $\alpha$ parametrizes the polynomial eigenvalue decay of the regular Neural Tangent Kernel (NTK), $\lambda_j\asymp j^{-2\alpha}$, and $2\alpha=d/(d-1)$ under the spherical uniform design considered here. The target function is assumed to belong to the interpolation space $[\mathcal H_K]^{s'}$ with $s'\ge3$. This rate is even sharper than the currently known nearly-optimal rate of $\mathcal O(n^{-\frac{2\alpha s'}{2\alpha s'+1}})\log^2(1/\delta)$, where $n$ is the size of the training data and $\delta \in (0,1)$ is a small probability. This rate is also sharper than the standard kernel regression rate of $\mathcal O(n^{-\frac{2\alpha}{2\alpha+1}})$ obtained under the regular Neural Tangent Kernel (NTK) regime when training the neural network with the vanilla gradient descent (GD), where $2\alpha = d/(d-1)$. Our analysis is based on two key technical contributions. First, we present a principled decom…
Learning with Importance-weighted Variational Inference
Several variational bounds involving importance weighting ideas generalize the Evidence Lower BOund (ELBO) for marginal likelihood optimization, such as the Importance-weighted Auto-Encoder (IWAE), Variational Rényi (VR) and VR-IWAE bounds. Yet, it remains unclear how the joint choice of bound and gradient estimator impacts the behavior of the resulting variational inference (VI) algorithms. This paper provides a unified theoretical comparison of reparameterized (REP) and doubly-reparameterized (DREP) gradient estimators tied to the IWAE, VR and VR-IWAE bounds. Through asymptotic analyses of the Signal-to-Noise Ratio as the number of Monte Carlo samples $N$ goes to infinity, we identify a bias-variance tradeoff in these gradient estimators and we formally justify the superiority of DREP over REP in importance-weighted VI. An additional asymptotic analysis for challenging regimes, where both $N$ and the Kullback-Leibler divergence between the variational and posterior densities go to infinity, indicates that importance-weighted VI gradient estimators point in a well-founded direction even when the variational approximation deteriorates. Together, these complementary results characterize the optimization trajectory in importance-weighted VI from poor initialization to final convergence. Importantly, our proof techniques establish general theoretical tools for the study of sample means ratios whose scope extends beyond VI and constitute an independent contribution to the field…
Discovering and Reasoning of Causality in the Hidden World with Large Language Models
Revealing hidden causal variables alongside the underlying causal mechanisms is essential to the development of science. Despite the progress in the past decades, existing practice in causal discovery (CD) heavily relies on high-quality measured variables, which are usually given by human experts. In fact, the lack of well-defined high-level variables behind unstructured data has been a longstanding roadblock to a broader real-world application of CD. This procedure can naturally benefit from an automated process that can suggest potential hidden variables in the system. Interestingly, Large language models (LLMs) are trained on massive observations of the world and have demonstrated great capability in processing unstructured data. To leverage the power of LLMs, we develop a new framework termed Causal representation Assistant (COAT) that incorporates the rich world knowledge of LLMs to propose useful measured variables for CD with respect to high-value target variables on their paired unstructured data. Instead of directly inferring causality with LLMs, COAT constructs feedback from intermediate CD results to LLMs to refine the proposed variables. Given the target variable and the paired unstructured data, we first develop COAT-MB that leverages the predictivity of the proposed variables to iteratively uncover the Markov Blanket of the target variable. Built upon COAT-MB, COAT-PAG further extends to uncover a more complete causal graph, i.e., Partial Ancestral Graph, by it…
Alternating Regret for Online Convex Optimization
Motivated by alternating learning dynamics in two-player games, a recent work by Cevher et al. (2023) shows that $o(\sqrt{T})$ alternating regret is possible for any $T$-round adversarial Online Linear Optimization (OLO) problem, and left as an open question whether the same is true for general Online Convex Optimization (OCO). We answer this question in the affirmative by showing that the continuous Hedge algorithm achieves $\tilde{\mathcal{O}}(d^{\frac{2}{3}}T^{\frac{1}{3}})$ alternating regret for any adversarial $d$-dimensional OCO problems. We show that this implies an alternating learning dynamic that finds a Nash equilibrium for any convex-concave zero-sum games or a coarse correlated equilibrium for any convex two-player general-sum games at a rate of $\tilde{\mathcal{O}}(d^{\frac{2}{3}}/T^{\frac{2}{3}})$. To further improve the time complexity and/or the dimension dependence, we propose another simple algorithm, Follow-the-Regularized-Leader with a regularizer whose convex conjugate is 3rd-order smooth, for OCO with smooth and self-concordant loss functions (such as linear or quadratic losses). We instantiate our algorithm with different regularizers and show that, for example, when the decision set is the $\ell_2$ ball, our algorithm achieves $\tilde{\mathcal{O}}(T^{\frac{2}{5}})$ alternating regret with no dimension dependence (and a better $\tilde{\mathcal{O}}(T^{\frac{1}{3}})$ bound for quadratic losses). We complement our results by showing some algorithm-speci…
Flexible Modeling of Information Diffusion on Networks with Statistical Guarantees
Modeling information spread through a network is one of the key problems of network analysis. Most approaches assume the spread is governed by some probabilistic diffusion model, often parameterized by the strength of connections between network members (edge weights), highlighting the need for methods that can accurately estimate them. Multiple prior works suggest estimators for particular diffusion models, but lack a rigorous statistical analysis establishing their asymptotic properties and allowing for uncertainty quantification. In this paper, we develop a likelihood-based approach to estimate edge weights from the observed information diffusion paths under the proposed General Linear Threshold (GLT) model, a broad class of discrete-time information diffusion models that includes both the well-known linear threshold (LT) and independent cascade (IC) models. We derive a finite-sample error bound for the estimator and demonstrate that it is asymptotically normal under mild conditions. We conclude by studying the GLT model in the context of the Influence Maximization (IM) problem. Extensive experiments on synthetic and real-world networks demonstrate that the proposed estimation and inference frameworks for the GLT model parameters can significantly improve estimation of spread from a given subset of nodes, prediction of node activation, and the quality of the IM problem solutions.
Provable Hierarchical Martingale-based Testing for Visual Anomaly Detection
Existing visual anomaly detection approaches primarily focus on designing robust score functions but often neglect investigating the decision-making process, which plays a key role in reliable and safety-critical applications. Currently, the decision-making process suffers from two key challenges: distribution shift and theoretical guarantee. To address these issues, this paper proposes a novel approach called Hierarchical Martingale-based Testing framework (HiMT). Specifically, our approach leverages martingale theory to mitigate distribution shift issue and employs a hierarchical multiple testing procedure to rigorously control decision errors at both image level and pixel level. Extensive experimental results verify the superior decision-making capabilities of our proposed HiMT framework at both image level and pixel level.
Stochastic Bilevel Optimization with Heavy-Tailed Noise
This paper considers the smooth bilevel optimization in which the lower-level problem is strongly convex and the upper-level problem is possibly nonconvex. We focus on the stochastic setting where the algorithm can access the unbiased stochastic gradient evaluation with heavy-tailed noise, which is prevalent in many machine learning applications, such as training large language models and reinforcement learning. We propose a nested-loop normalized stochastic bilevel approximation (N$^2$SBA) for finding an $\epsilon$-stationary point with the stochastic first-order oracle (SFO) complexity of $\tilde{\mathcal{O}}\big(\kappa^{\frac{7p-3}{p-1}} \sigma^{\frac{p}{p-1}} \epsilon^{-\frac{4 p - 2}{p-1}}\big)$, where $\kappa$ is the condition number, $p\in(1,2]$ is the order of central moment for the noise, and $\sigma$ is the noise level. Furthermore, we specialize our idea to solve the nonconvex-strongly-concave minimax optimization problem, achieving an $\epsilon$-stationary point with the SFO complexity of $\tilde{\mathcal O}\big(\kappa^{\frac{2p-1}{p-1}} \sigma^{\frac{p}{p-1}} \epsilon^{-\frac{3p-2}{p-1}}\big)$. All the above upper bounds match the best-known results under the special case of the bounded variance setting, i.e., $p=2$. We also conduct the numerical experiments to show the empirical superiority of proposed methods.
Anderson Mixing in Bures–Wasserstein Space of Gaussian Measures
Various statistical tasks, including sampling or computing Wasserstein barycenters, can be reformulated as fixed-point problems for operators on probability distributions. Accelerating standard fixed-point iteration schemes provides a promising novel approach to the design of efficient numerical methods for these problems. The Wasserstein geometry on the space of probability measures, although not precisely Riemannian, allows us to define various useful Riemannian notions, such as tangent spaces, exponential maps and parallel transport, motivating the adaptation of Riemannian numerical methods. We demonstrate this by developing and implementing the Riemannian Anderson mixing (RAM) method for Gaussian distributions. The method reuses the history of the residuals and improves the iteration complexity, and we argue that the additional costs, compared to the Picard method, are negligible. We show that certain balls in the Bures--Wasserstein manifold satisfy the requirements for convergence of RAM. The numerical experiments show a significant acceleration compared to the Picard iteration, and performance on par with Riemannian gradient descent and conjugate gradient methods.
A New Family of Poisson Non-negative Matrix Factorization Methods Using the Shifted Log Link
Poisson non-negative matrix factorization (NMF) is a widely used method to find interpretable "parts-based" decompositions of count data. While many variants of Poisson NMF exist, they all assume that the "parts" in the decomposition combine additively. This assumption may be natural in some settings, but not in others. Here, we introduce Poisson NMF with the shifted-log link function to relax this assumption. The shifted-log link function has a single tuning parameter, and as this parameter varies the model changes from assuming that parts combine additively to assuming that parts combine more multiplicatively. We provide an algorithm to fit this model by maximum likelihood, and also an approximation that substantially reduces computation time for large, sparse datasets. We illustrate these new methods on a variety of real and simulated data. Our examples show that while standard Poisson NMF often leads to sparse, cluster-like representations of data, the shifted-log link can produce more modular, layered results.
Bayesian Predictive Inference Beyond Martingales
There is growing interest in the so-called Bayesian Predictive Inference approach, which enables Bayesian inference without specifying a likelihood function or prior distribution and without resorting to MCMC methods. Instead, only a sequence of predictive distributions for the observations is required, and inference on the unknown estimand can be performed, cheaply in parallel, using bootstrap-type schemes. Understanding which classes of predictive distributions can be used within this framework, is still a key open question. We relax commonly used probabilistic assumptions on the observations, namely exchangeability and conditionally identical distribution, and on their predictive distributions, which are measure-valued martingales, by introducing the new class of Almost Conditionally Identically Distributed (a.c.i.d.) random variables. This class assumes that the predictive distributions are measure-valued almost supermartingales, and is parametrized by a sequence of parameters $(\xi_{n})_{n\geq 0}$, which regulate the decay of conditional dependence among future observations. Under mild summability assumptions on $(\xi_{n})_{n\geq 0}$, the resulting sequence of observations is shown to be asymptotically exchangeable, hence amenable to Bayesian Predictive Inference techniques. A.c.i.d. random variables arise naturally in recursive algorithms, and include classic approaches in Statistics and Learning Theory, such as kernel estimators, and more novel ones, such as the param…
CLASSIX: A High-Performance Library for Explainable~Clustering
We present CLASSIX, an open-source Python library for fast and explainable clustering. The library provides a unified interface to three metric-specific implementations, compiled acceleration with portable Python fallbacks, and extensive user and developer documentation. CLASSIX accepts NumPy arrays and Pandas DataFrames and is available under the MIT license at https://github.com/nla-group/classix.
Sample-Efficient Regret-Minimizing Double Oracle in Two-Player Zero-Sum Extensive-Form Games
Extensive-Form Game (EFG) represents a fundamental model for analyzing sequential interactions among multiple agents and the primary challenge to solve it lies in mitigating sample complexity. Existing research indicates that Double Oracle (DO) can reduce the sample complexity dependence on the information set number $|\mathcal{S}|$ to a restricted game size $X$ in solving two-player zero-sum EFGs. This is attributed to the early convergence of full-game Nash Equilibrium (NE) through iteratively solving restricted games. However, our analysis of Extensive-Form Double Oracle (XDO) yields a sample-complexity upper bound exponential in the number of restricted games, due to its exponentially increasing restricted game expansion frequency. Here we introduce Adaptive Double Oracle (AdaDO) to significantly alleviate sample complexity to polynomial by deploying the optimal expansion frequency. Furthermore, to comprehensively study the principles and influencing factors underlying sample complexity of DO, we introduce a novel theoretical framework Regret-Minimizing Double Oracle (RMDO) to provide directions for designing efficient DO algorithms. Empirical results demonstrate that AdaDO attains the superior approximation of NE with less sample complexity than the strong baselines including Linear CFR, MCCFR and existing DO. Importantly, combining RMDO with warm starting and stochastic regret minimization further improves convergence rate and scalability, thereby paving the way for ad…
Semidefinite programming relaxations and debiasing for MAXCUT-based clustering
In this paper, we consider the problem of partitioning a small data sample of size $n$ drawn from a mixture of $2$ subgaussian distributions in $\mathbb{R}^p$. We consider semidefinite programming relaxations of an integer quadratic program that is formulated essentially as finding the maximum cut on a graph, where edge weights in the cut represent dissimilarity scores between two nodes based on their $p$ features. We define the signal-to-noise ratio (SNR) as $s^2 := \min\{n p \gamma^2, \Delta^2\}$, where $\Delta^2 := p \gamma$ denotes the $\ell_2^2$ distance between the two cluster centers. For both balanced and unbalanced cases, we allow each population to have distinct covariance structures $\Sigma_1, \Sigma_2$ with diagonal matrices as special cases. Our contributions are twofold. First, we provide a unified framework for analyzing three computationally efficient algorithms: SDP1, BalancedSDP, and Spectral clustering, yielding universal polynomial-rate misclassification guarantees for all three algorithms. Moreover, our framework accom-modates a full range of tradeoffs between $n, p,$ and $\gamma$, allowing for partial recovery (success rate $\lt 100\%$) when $s^2$ is lower bounded by a constant. Second, we prove that the misclassification errors for SDP1 and BalancedSDP decay exponentially with respect to the SNR $s^2$ and the BalancedSDP requires no explicit debiasing when the two clusters have equal sizes. To our knowledge, this is the first time such results are obta…
Anthropic's Claude Code Lets Developers Theme Their AI Coding Desktop
Anthropic rolled out custom theming for Claude Code in the desktop app, letting you restyle colors and drop in animated backdrops through mods.
Photos and video show a cruise ship sinking after it was struck by a tanker in New Orleans
A cruise ship docked on the Mississippi River in New Orleans collided with a tanker on Friday.