This paper explores whether compact quantum-inspired recurrent models can efficiently forecast full network traffic matrices under realistic resource constraints. The authors adapt gated quantum-inspired Kolmogorov–Arnold Network Fast-Weight Programmers, or QKAN-FWPs, to predict the next 20 five-minute Abilene traffic-matrix frames from a two-hour history, covering 144 origin–destination traffic channels. Instead of relying on graph, transformer, vision, or diffusion modules, the study isolates lightweight recurrent fast-weight designs and compares three QKAN placement variants against matched-size and larger LSTMs plus a classical gated FWP baseline. The best model, G-QKANFWP, achieves the lowest pooled RMSE while using only 22.4% of the parameters of the larger LSTM, and it also outperforms the classical G-FWP baseline, suggesting that the quantum-inspired fast readout contributes beyond the gated fast-weight mechanism alone. The results position G-QKANFWP as a promising accuracy–efficiency trade-off for online traffic-matrix forecasting in edge, cloud-edge, and network-control settings.</p>\n","updatedAt":"2026-07-03T17:35:05.661Z","author":{"_id":"68b93b3a6c86c127a199ad90","avatarUrl":"/avatars/f370d99b240ce5b9e1bfdbd3130d9ee4.svg","fullname":"Jiun-Cheng Jiang","name":"Jim137","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isUserFollowing":false,"primaryOrg":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/65df9200dc3292a8983e5017/Vs5FPVCH-VZBipV3qKTuy.png","fullname":"NVIDIA","name":"nvidia","type":"org","isHf":false,"plan":"plus"}}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.8472991585731506},"editors":["Jim137"],"editorAvatarUrls":["/avatars/f370d99b240ce5b9e1bfdbd3130d9ee4.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2606.27821","authors":[{"_id":"6a47f15b3daa34221c7c1c67","name":"Kuo-Chung Peng","hidden":false},{"_id":"6a47f15b3daa34221c7c1c68","name":"Jiun-Cheng Jiang","hidden":false},{"_id":"6a47f15b3daa34221c7c1c69","name":"Chun-Hua Lin","hidden":false},{"_id":"6a47f15b3daa34221c7c1c6a","name":"Tai-Yue Li","hidden":false},{"_id":"6a47f15b3daa34221c7c1c6b","name":"Nan-Yow Chen","hidden":false},{"_id":"6a47f15b3daa34221c7c1c6c","name":"Samuel Yen-Chi Chen","hidden":false}],"mediaUrls":["https://cdn-uploads.huggingface.co/production/uploads/68b93b3a6c86c127a199ad90/mWDu7SCF6i0b4Oxtp7rxd.png"],"publishedAt":"2026-06-26T00:00:00.000Z","submittedOnDailyAt":"2026-07-03T00:00:00.000Z","title":"Parameter-Efficient Quantum-Inspired Fast Weight Programmers for Traffic-Matrix Forecasting","submittedOnDailyBy":{"_id":"68b93b3a6c86c127a199ad90","avatarUrl":"/avatars/f370d99b240ce5b9e1bfdbd3130d9ee4.svg","isPro":false,"fullname":"Jiun-Cheng Jiang","user":"Jim137","type":"user","name":"Jim137"},"summary":"Traffic matrices (TMs) capture network-wide origin-destination demand and are central to traffic engineering, yet accurate whole-matrix forecasting remains challenging when prediction must be performed under the memory, update, and training-budget constraints of online network control. This paper investigates whether compact quantum-inspired recurrent models can provide effective TM forecasts without relying on dedicated graph, transformer, or diffusion modules. We adapt gated quantum-inspired Kolmogorov-Arnold network fast-weight programmers (QKAN-FWPs) to direct multi-step Abilene TM forecasting, where each model predicts the next 20 five-minute frames of a 144-channel origin-destination (OD) matrix from a two-hour history. We benchmark three QKAN placement variants against a matched-size long short-term memory (LSTM) network, a larger LSTM, and a classical gated fast-weight programmer under a shared fixed-budget training protocol. Among the evaluated recurrent models, G-QKANFWP achieves the best pooled root-mean-square error (RMSE), while using only 22.4% of the larger LSTM. It also outperforms both the matched-size LSTM and the classical G-FWP baseline, indicating that the gain is not due to gated fast-weight framework alone. Convergence and channel-wise analyses further show that the quantum-inspired variants obtain lower validation-loss area under the learning curve (AULC) than matched-size recurrent baselines, while G-QKANFWP and GQKAN-FWP achieve substantially more OD-channel wins. These results identify a classical slow programmer with a quantum-inspired fast programmer as a promising accuracy-efficiency design for resource-conscious network traffic-matrix forecasting.","upvotes":1,"discussionId":"6a47f15c3daa34221c7c1c6d","ai_summary":"Quantum-inspired recurrent models using gated QKAN-FWPs demonstrate superior forecasting accuracy with reduced computational requirements compared to traditional LSTM networks for traffic matrix prediction.","ai_keywords":["traffic matrices","quantum-inspired recurrent models","gated quantum-inspired Kolmogorov-Arnold network fast-weight programmers","QKAN-FWPs","multi-step forecasting","origin-destination matrix","long short-term memory","LSTM","classical gated fast-weight programmer","root-mean-square error","validation-loss area under learning curve","AULC"],"ai_summary_model":"Qwen/Qwen2.5-Coder-32B-Instruct"},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"68b93b3a6c86c127a199ad90","avatarUrl":"/avatars/f370d99b240ce5b9e1bfdbd3130d9ee4.svg","isPro":false,"fullname":"Jiun-Cheng Jiang","user":"Jim137","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2606/2606.27821.md","query":{}}">

Get this paper in your agent:

Don't have the latest CLI?

Cite arxiv.org/abs/2606.27821 in a model README.md to link it from this page.

Cite arxiv.org/abs/2606.27821 in a dataset README.md to link it from this page.

Cite arxiv.org/abs/2606.27821 in a Space README.md to link it from this page.