digitado

Cooperation in multi-agent reinforcement learning (MARL) benefits from inter-agent communication, yet most approaches assume idealized channels and existing value decomposition methods ignore who successfully shared information with whom. We propose CLOVER, a cooperative MARL framework whose centralized value mixer is conditioned on the communication graph realized under a realistic wireless channel. This graph introduces a relational inductive bias into value decomposition, constraining how individual utilities are mixed based on the realized communication structure. The mixer is a GNN […]

Ver mais

The disintegration of liquid sheets into ligaments and droplets involves highly transient, multi-scale dynamics that are difficult to quantify from high-speed shadowgraphy images. Identifying droplets, ligaments, and blobs formed during breakup, along with tracking across frames, is essential for spray analysis. However, conventional multi-object tracking frameworks impose strict one-to-one temporal associations and cannot represent one-to-many fragmentation events. In this study, we present a two-stage deep learning framework for object detection and temporal relationship modeling across frames. The framework […]

Ver mais

Recently posted some info on a full “intelligence engine” we’ve been working on. reinforcement learning framework that uses Q-learning with entropy-based exploration control to classify structured datasets. I’ve been running it across multiple domains and just released the datasets publicly. The most interesting one: I ran it against the entire OEIS (Online Encyclopedia of Integer Sequences) — 390,952 sequences. The agent classifies each sequence by information-theoretic properties: Shannon entropy of term values, growth dynamics, periodicity, convergence behavior, and […]

Ver mais

Visual decoding from brain signals is a key challenge at the intersection of computer vision and neuroscience, requiring methods that bridge neural representations and computational models of vision. A field-wide goal is to achieve generalizable, cross-subject models. A major obstacle towards this goal is the substantial variability in neural representations across individuals, which has so far required training bespoke models or fine-tuning separately for each subject. To address this challenge, we introduce a meta-optimized approach for semantic visual […]

Ver mais

Saw a post on X saying that the latitude of the Pyramid of Giza is the same as the speed of light. I looked into this, expecting it to be approximately true. It’s exactly true in the sense that the speed of light in vacuum is 299,792,458 m/s and the line of latitude 29.9792458° N passes through the pyramid. The exact center of the pyramid is at 29.97917° N, 31.13417° E. Of course this is a coincidence. Even […]

Ver mais