SERMAYE KISITLI YAPAY ZEKA YATIRIMI: DİL MODELİ İNCE AYARI İÇİN BİR GERÇEK OPSİYON VE VERİMLİLİK SINIRI YAKLAŞIMI

Ertunç ERERDİ; Onur ALTUNTAŞ; Tuba ÇELİK; Mücahit AKIN

doi:10.64734/bjss.2-1-02

Yazarlar

Ertunç ERERDİ MAD CAT LABS Author
Onur ALTUNTAŞ MAD CAT LABS Author
Tuba ÇELİK MAD CAT LABS Author
Mücahit AKIN MAD CAT LABS Author

DOI:

https://doi.org/10.64734/bjss.2-1-02

Anahtar Kelimeler:

gerçek opsiyonlar- sermaye bütçelemesi- verimlilik sınırı- yapay zekâ yatırımı- dil modeli ince ayarı

Öz

Büyük dil modelleri (BDM) üzerine yapılan ince ayar yatırımları, son yıllarda kurumların yapay zeka stratejilerinin merkezine yerleşmiştir. Bu çalışma, ince ayar konfigürasyonu seçimini salt teknik bir tercih olarak değil, sermaye kısıtlı ortamlarda geri dönülemez yatırım kararı olarak modellemekte ve gerçek opsiyonlar teorisi ile portföy verimlilik sınırı çerçevesini entegre eden bir karar mimarisi önermektedir. Önermeler, Qwen3.5 model ailesi üzerinde dört eksiksiz ince ayar koşusuyla ampirik olarak illüstre edilmiştir. Bulgular, Qwen3.5-9B konfigürasyonunun verimlilik sınırı üzerinde optimal durdurma noktasını oluşturduğunu göstermektedir.

İndirmeler

İndirme verisi henüz mevcut değil.

Referanslar

Aryan, V., Bertolini, M., & Butler, P. (2023). Capital budgeting under technological uncertainty: A real options perspective on AI infrastructure investments. Long Range Planning, 56(4), 102345.

Brealey, R. A., Myers, S. C., & Allen, F. (2020). Principles of corporate finance (13th ed.). McGraw-Hill.

Brown, T., Mann, B., Ryder, N., et al. (2020). Language models are few-shot learners. Advances in Neural Information Processing Systems, 33, 1877-1901.

Chen, T., Xu, B., Zhang, C., & Guestrin, C. (2016). Training deep nets with sublinear memory cost. arXiv:1604.06174.

Dettmers, T., Lewis, M., Shleifer, S., & Zettlemoyer, L. (2022). 8-bit optimizers via block-wise quantization. International Conference on Learning Representations.

Dettmers, T., Pagnoni, A., Holtzman, A., & Zettlemoyer, L. (2023). QLoRA: Efficient finetuning of quantized LLMs. Advances in Neural Information Processing Systems, 36.

Dixit, A. K., & Pindyck, R. S. (1994). Investment under uncertainty. Princeton University Press.

González, M., López, A., & Martínez, R. (2026). Pareto frontier analysis for cost-efficient fine-tuning of large language models. Decision Support Systems, forthcoming.

Hu, E. J., Shen, Y., Wallis, P., et al. (2021). LoRA: Low-rank adaptation of large language models. International Conference on Learning Representations.

Lorie, J. H., & Savage, L. J. (1955). Three problems in rationing capital. Journal of Business, 28(4), 229-239.

Markowitz, H. (1952). Portfolio selection. Journal of Finance, 7(1), 77-91.

McDonald, R., & Siegel, D. (1986). The value of waiting to invest. Quarterly Journal of Economics, 101(4), 707-728.

Mishra, S., Ewing, M. T., & Cooper, H. B. (2022). Artificial intelligence focus and firm performance. Journal of the Academy of Marketing Science, 50, 1176-1197.

Pan, K., Liu, J., & Wang, M. (2025). Break-even analysis of on-premise versus cloud-based deployment of large language models. Journal of Management Information Systems, 42(1), 88-117.

Patterson, D., Gonzalez, J., Le, Q., Liang, C., Munguia, L., Rothchild, D., So, D., Texier, M. & Dean, J. (2021). Carbon emissions and large neural network training. arXiv:2104.10350.

Ross, S. A., Westerfield, R. W., & Jaffe, J. (2019). Corporate finance (12th ed.). McGraw-Hill.

Strubell, E., Ganesh, A., & McCallum, A. (2019). Energy and policy considerations for deep learning in NLP. Annual Meeting of the Association for Computational Linguistics, 3645-3650.

Trigeorgis, L. (1996). Real options: Managerial flexibility and strategy in resource allocation. MIT Press.

Vaswani, A., Shazeer, N., Parmar, N.,Uszkoreit, J., Jones, L., Gomez, A.N., Kaiser, L. & Polosukhin, I. (2017). Attention is all you need. Advances in Neural Information Processing Systems, 30.

Weingartner, H. M. (1963). Mathematical programming and the analysis of capital budgeting problems. Prentice-Hall.

Yang, A., Yang, B., Hui, B., et al. (2024). Qwen2.5 technical report. arXiv:2412.15115.

Yuan, J., Zhang, Y., Wang, X., & Liu, F. (2025). No free lunch in memory-efficient fine-tuning: A multi-axis empirical analysis. Information Systems Research, 36(2), 412-438.