technocracy

We study a canonical multi-task demand learning problem motivated by retail pricing, in which a firm seeks to estimate heterogeneous linear price-response functions across a large collection of decision contexts. Each context is characterized by rich observable covariates yet typically exhibits only limited historical price variation, motivating the use of multi-task learning to borrow strength across tasks. A central challenge in this setting is endogeneity: historical prices are chosen by managers or algorithms and may be arbitrarily correlated […]

Ver mais

Training relational foundation models requires learning representations that transfer across tasks, yet available supervision is typically limited to a small number of prediction targets per database. This task scarcity causes learned representations to encode task-specific shortcuts that degrade transfer even within the same schema, a problem we call label leakage. We study this using K-Space, a modular architecture combining frozen pretrained tabular encoders with a lightweight message-passing core. To suppress leakage, we introduce a gradient projection method that […]

Ver mais

arXiv:2601.09727v1 Announce Type: new Abstract: Cross-domain scientific synthesis requires connecting mechanistic explanations across fragmented literature, a capability that remains challenging for both retrieval-based systems and unconstrained language models. While recent work has applied large language models to scientific summarization and question answering, these approaches provide limited control over reasoning depth and structural grounding. We frame mechanistic synthesis as a graph-constrained multi-hop reasoning problem over literature-derived concept graphs. Given a scientific query and a compact, query-local corpus, SciNets constructs […]

Ver mais

Now, if your CEO has never heard the phrase Ralph Loop, oh man, you are less than 30 days away from your next promotion. I’m not even exaggerating. Walk into his office, close the door, and say, hey chief, been experimenting with something. It’s called Ralph Loops. And I think it could change literally everything. And he’s gonna say, what’s a Ralph loop? And you will say, give me $18,000 worth of API credits and I’ll show you. […]

Ver mais

Author(s): Dr Swarneendu AI Originally published on Towards AI. There are next-word predictions your model is mathematically forbidden from making. Not unlikely. Forbidden, the way a piano with too few keys cannot play a note that lies past its keyboard. The proof needs nothing but small whole numbers and patience. We will do every step on paper. Summary of the article Read the full blog for free on Medium. Join thousands of data leaders on the AI newsletter. […]

Ver mais

Hybrid architectures combining state-space models with attention have achieved strong efficiency-quality tradeoffs, yet existing approaches either apply attention uniformly or learn static sparse patterns. This misses a key opportunity: emph{attention demand should decrease over time as recurring patterns become familiar}. We present a surprising finding from analyzing GPT-2 models: textbf{88%} of attention operations retrieve information already predictable from the model’s hidden state, and this redundancy does emph{not} decrease during training. Motivated by this observation, we introduce textbf{ours{}} (textbf{C}onsolidation-based […]

Ver mais

arXiv:2602.05079v1 Announce Type: new Abstract: The problem with existing camera-based Deep Reinforcement Learning approaches is twofold: they rarely integrate high-level scene context into the feature representation, and they rely on rigid, fixed reward functions. To address these challenges, this paper proposes a novel pipeline that produces a neuro-symbolic feature representation that encompasses semantic, spatial, and shape information, as well as spatially boosted features of dynamic entities in the scene, with an emphasis on safety-critical road users. It also […]

Ver mais

Neural networks are central to modern artificial intelligence, yet their training remains highly sensitive to data contamination. Standard neural classifiers are trained by minimizing the categorical cross-entropy loss, corresponding to maximum likelihood estimation under a multinomial model. While statistically efficient under ideal conditions, this approach is highly vulnerable to contaminated observations including label noises corrupting supervision in the output space, and adversarial perturbations inducing worst-case deviations in the input space. In this paper, we propose a unified and […]

Ver mais

arXiv:2603.17381v3 Announce Type: replace-cross Abstract: AI coding agents make empirical specification search fast and cheap, but they also widen hidden researcher degrees of freedom. Building on an open-source agent-loop architecture, this paper adapts that framework to an empirical economics workflow and adds a post-search holdout evaluation. In a forecast-combination illustration, multiple independent agent runs outperform standard benchmarks in the original rolling evaluation, but not all continue to do so on a post-search holdout. Logged search and holdout evaluation […]

Ver mais