Artificial Intelligence and Machine Learning ResearchBlog by Marek Barakhttps://n1o.github.io/RL Bite: Monte Carlo Search Treehttps://n1o.github.io/posts/rl-bite-monte-carlo-search-tree/https://n1o.github.io/posts/rl-bite-monte-carlo-search-tree/Thu, 10 Apr 2025 07:00:56 GMTRL Bite: Monotonic Policy Improvement and Deriving Proximal Policy Optimization (PPO)https://n1o.github.io/posts/rl-bite-policy-improvement/https://n1o.github.io/posts/rl-bite-policy-improvement/Tue, 01 Apr 2025 06:44:11 GMTRL Bite: Policy Gradient and Reinforcehttps://n1o.github.io/posts/rl-bite-policy-gradient-and-reinforce/https://n1o.github.io/posts/rl-bite-policy-gradient-and-reinforce/Sat, 08 Mar 2025 12:24:13 GMTRL Bite: Learning the Q Functionhttps://n1o.github.io/posts/rl-bite-learning-the-q-function/https://n1o.github.io/posts/rl-bite-learning-the-q-function/Mon, 03 Mar 2025 07:59:26 GMTTLDR; Graph Contrastive Learning: Representation Scatteringhttps://n1o.github.io/posts/tldr-graph-contrastive-repr-representation-shattering/https://n1o.github.io/posts/tldr-graph-contrastive-repr-representation-shattering/Sun, 23 Feb 2025 12:04:42 GMTRL Bite: Computing the Value Functionhttps://n1o.github.io/posts/rl-bite-computing-value-functions/https://n1o.github.io/posts/rl-bite-computing-value-functions/Tue, 18 Feb 2025 05:20:35 GMTTLDR; HC-GAE The Hierarchical Cluster-based Graph Auto-Encoder for Graph Representation Learninghttps://n1o.github.io/posts/tldr-hc-gae/https://n1o.github.io/posts/tldr-hc-gae/Sun, 16 Feb 2025 13:48:13 GMTRL Bite: Bellmans Equations and Value Functionshttps://n1o.github.io/posts/rl-bite-bellmans-equations-and-value-functions/https://n1o.github.io/posts/rl-bite-bellmans-equations-and-value-functions/Wed, 12 Feb 2025 06:00:52 GMTTLDR; Duplex: Dual GAT for Complex Embeddings of Directed Graphshttps://n1o.github.io/posts/tldr-duplex/https://n1o.github.io/posts/tldr-duplex/Mon, 10 Feb 2025 08:22:15 GMTRL Bite: Exploitation vs Explorationhttps://n1o.github.io/posts/rl-bite-exploration-vs-exploitation/https://n1o.github.io/posts/rl-bite-exploration-vs-exploitation/Wed, 05 Feb 2025 08:51:24 GMTGraph Neural Networks meet Large Language Modelshttps://n1o.github.io/posts/graph-neural-networks-meet-large-language-models/https://n1o.github.io/posts/graph-neural-networks-meet-large-language-models/Mon, 16 Dec 2024 08:17:05 GMTHymba, a new breed of SSM-Attention Hybridshttps://n1o.github.io/posts/hymba-new-ssm-att-hybrid-breed/https://n1o.github.io/posts/hymba-new-ssm-att-hybrid-breed/Mon, 02 Dec 2024 06:00:00 GMTTransform any LLMs to a powerful Encoderhttps://n1o.github.io/posts/from-llms-to-sentence-embeddings/https://n1o.github.io/posts/from-llms-to-sentence-embeddings/Mon, 25 Nov 2024 08:00:33 GMT2025 Year of Zighttps://n1o.github.io/posts/2025-yearl-of-zig/https://n1o.github.io/posts/2025-yearl-of-zig/Wed, 20 Nov 2024 07:31:40 GMTDistilling State Space Models from Transformershttps://n1o.github.io/posts/distilling-ssm-from-transformers/https://n1o.github.io/posts/distilling-ssm-from-transformers/Mon, 28 Oct 2024 09:12:27 GMTIllusion of State in SSMs like Mambahttps://n1o.github.io/posts/ssm-the-illusion/https://n1o.github.io/posts/ssm-the-illusion/Mon, 14 Oct 2024 08:52:09 GMTMamba(2) and Transformer Hybrids: An Overviewhttps://n1o.github.io/posts/ssm-transformer-hybrids-guide/https://n1o.github.io/posts/ssm-transformer-hybrids-guide/Wed, 18 Sep 2024 09:26:25 GMTHydra a Double Headed Mambahttps://n1o.github.io/posts/hydra-a-double-headed-mamba/https://n1o.github.io/posts/hydra-a-double-headed-mamba/Thu, 29 Aug 2024 09:53:51 GMTFrom Mamba to Mamba-2https://n1o.github.io/posts/from-mamba-to-mamba2/https://n1o.github.io/posts/from-mamba-to-mamba2/Thu, 08 Aug 2024 07:57:32 GMTButterflies, Monarchs, Hyenas, and Lightning Fast BERThttps://n1o.github.io/posts/butterflies-monarchs-hyenas-and-lightning-fast-bert/https://n1o.github.io/posts/butterflies-monarchs-hyenas-and-lightning-fast-bert/Fri, 12 Jul 2024 11:36:49 GMTBinT5 and HexT5 or T5 and Binary Reverse Engineeringhttps://n1o.github.io/posts/t5-and-reverse-engineering/https://n1o.github.io/posts/t5-and-reverse-engineering/Wed, 12 Jun 2024 12:07:17 GMTCodeT5 and CodeT5+https://n1o.github.io/posts/code-t5-plus/https://n1o.github.io/posts/code-t5-plus/Sat, 01 Jun 2024 06:46:43 GMTLonger Context for T5https://n1o.github.io/posts/longer-context-for-t5/https://n1o.github.io/posts/longer-context-for-t5/Mon, 29 Apr 2024 12:10:36 GMTT5 the Old New Thinghttps://n1o.github.io/posts/t5-the-old-new-thing/https://n1o.github.io/posts/t5-the-old-new-thing/Wed, 06 Mar 2024 11:58:32 GMTNixos for Hobby Projecthttps://n1o.github.io/posts/nixos-for-hobby-project/https://n1o.github.io/posts/nixos-for-hobby-project/Wed, 13 Sep 2023 11:09:10 GMTPaper overview: Hungry Hungry Hippos: Towards Language Modeling with State Space Modelshttps://n1o.github.io/posts/hungry-hungry-hippos/https://n1o.github.io/posts/hungry-hungry-hippos/Mon, 06 Feb 2023 08:39:03 GMTPaper overview: Continuous-Time Modeling of Counterfactual Outcomes Using Neural Controlled Differential Equationshttps://n1o.github.io/posts/continuous-time-modeling-of-counterfatual-outcomes/https://n1o.github.io/posts/continuous-time-modeling-of-counterfatual-outcomes/Mon, 16 Jan 2023 16:37:36 GMTHierarchical Probabilistic Matrix Factorizationhttps://n1o.github.io/posts/pooled_matrix_factorization/https://n1o.github.io/posts/pooled_matrix_factorization/Hierarchical Probabilistic Matrix Factorization. An example implementation in Pyro.Thu, 17 Dec 2020 09:19:42 GMT