You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: class08/class08.jl
+18-14Lines changed: 18 additions & 14 deletions
Original file line number
Diff line number
Diff line change
@@ -41,7 +41,11 @@ References:
41
41
42
42
* Olfati-Saber, Reza, J. Alex Fax, and Richard M. Murray. "Consensus and cooperation in networked multi-agent systems." Proceedings of the IEEE 95.1 (2007): 215-233.
43
43
44
+
* Boyd, Stephen, et al. "Distributed optimization and statistical learning via the alternating direction method of multipliers." Foundations and Trends® in Machine learning 3.1 (2011): 1-122.
44
45
46
+
* Summers, Tyler H., and John Lygeros. "Distributed model predictive consensus via the alternating direction method of multipliers." 2012 50th annual Allerton conference on communication, control, and computing (Allerton). IEEE, 2012.
47
+
48
+
* Piansky, R., Stinchfield, G., Kody, A., Molzahn, D. K., & Watson, J. P. (2024). Long duration battery sizing, siting, and operation under wildfire risk using progressive hedging. Electric Power Systems Research, 235, 110785.
45
49
"""
46
50
47
51
# ╔═╡ 75bdf059-c8ac-4f9c-b023-3c010b4389cb
@@ -209,7 +213,7 @@ Defining the quadratic disagreement function $\varphi(x) = \frac{1}{2} x^T L x$,
209
213
"""
210
214
211
215
# ╔═╡ 46a8121f-f1aa-4d22-8ef9-1d02f957767d
212
-
question_box(md"""What motivates the division by 2 in the quadratic disagreement funcion?""")
216
+
question_box(md"""What motivates the division by 2 in the quadratic disagreement function?""")
213
217
214
218
# ╔═╡ 85b74a70-a8df-4741-aa1b-d551d0e9bea2
215
219
Foldable(md"Answer...", md"It accounts for the double-counting of undirected edges in the graph.")
@@ -779,7 +783,7 @@ We now consider a __more general form of consensus__, beyond the global consensu
779
783
$\sum_i f_i(x_i)$.
780
784
781
785
Each local variable $x_i$ contains a subset of components that correspond to certain components of a global variable $z$. We define a mapping $\mathcal{G}(i, j)$ from a local index $(i, j)$ to the corresponding global index $g$, such that
782
-
$(x_i)j = z{\mathcal{G}(i, j)}$.
786
+
$(x_i)_j = z_{\mathcal{G}(i, j)}$.
783
787
784
788
As a motivating example, consider a __model fitting__ problem. Here, the global variable $z$ represents the full feature vector, while each processor (or node) $i$ holds a subset of the data. The corresponding local variable $x_i$ represents the subset of features in $z$ that appear in block $i$ of the data. This structure is typical in large-scale, high-dimensional datasets that are __sparse__ and __distributed__ across multiple computing nodes.
785
789
@@ -1143,7 +1147,7 @@ In this chapter, we introduced __consensus__:
1143
1147
1144
1148
* a mechanism for __agreement__ among agents through local communication
1145
1149
* convergence guaranteed under __connectivity__ and __convexity__
1146
-
* enables coordination wtihout centralized control (scalable, robust)
1150
+
* enables coordination without centralized control (scalable, robust)
1147
1151
1148
1152
We also introduced __ADMM__, a powerful optimization framework:
1149
1153
@@ -1163,20 +1167,20 @@ question_box(md"""Consensus and ADMM rely on cooperation and information exchang
1163
1167
Foldable(md"Hint...", md"In such _adversarial_ or _strategic_ environments, the assumptions of standard consensus no longer hold. The optimization must be reformulated. It becomes __equilibrium-seeking__ rather than __agreement-seeking__.")
1164
1168
1165
1169
# ╔═╡ Cell order:
1166
-
#╠═462afff0-2ae3-4730-b48e-cf475fc9e14f
1167
-
#╠═195683a4-b093-46af-9fb2-0a8a67d996e8
1170
+
#╟─462afff0-2ae3-4730-b48e-cf475fc9e14f
1171
+
#╟─195683a4-b093-46af-9fb2-0a8a67d996e8
1168
1172
# ╟─75bdf059-c8ac-4f9c-b023-3c010b4389cb
1169
1173
# ╟─fe6b8381-edeb-4d0c-87f5-4ecd2b7b9183
1170
1174
# ╟─44154c9f-8e5e-49a7-b6ec-f465e0769c88
1171
-
#╠═40d12761-6ba5-4f92-aada-a26c9ddf5120
1175
+
#╟─40d12761-6ba5-4f92-aada-a26c9ddf5120
1172
1176
# ╟─242f3a4c-ad0b-4619-80e6-1f1ee244b6f4
1173
1177
# ╟─a6c12abf-4303-45bc-99a7-7b55061013d6
1174
1178
# ╟─cc94ceb2-e01a-4f10-9ee0-d0c906e5f46f
1175
1179
# ╟─453fb681-6f7b-42f2-9d10-0da7ad5b811c
1176
1180
# ╟─ee0fbf98-db44-4188-b623-3868a08c02b2
1177
-
#╠═46a8121f-f1aa-4d22-8ef9-1d02f957767d
1178
-
#╠═85b74a70-a8df-4741-aa1b-d551d0e9bea2
1179
-
#╠═524365a7-e799-4c55-acf4-88fcd5a716e2
1181
+
#╟─46a8121f-f1aa-4d22-8ef9-1d02f957767d
1182
+
#╟─85b74a70-a8df-4741-aa1b-d551d0e9bea2
1183
+
#╟─524365a7-e799-4c55-acf4-88fcd5a716e2
1180
1184
# ╟─3c5d6e70-6f43-11ef-3456-0123456789ab
1181
1185
# ╟─4d6e7f80-6f43-11ef-4567-0123456789ab
1182
1186
# ╟─5e7f8090-6f43-11ef-5678-0123456789ab
@@ -1209,15 +1213,15 @@ Foldable(md"Hint...", md"In such _adversarial_ or _strategic_ environments, the
1209
1213
# ╟─f75ec743-d448-432f-9511-b2c7a382873c
1210
1214
# ╟─b9146a5c-2898-4814-935a-6b1bfe7844c4
1211
1215
# ╟─aec156de-2fce-47bd-b5ad-11e7c62e2a74
1212
-
#╠═af74a473-4f23-45cc-883f-ed6ebe7167cf
1216
+
#╟─af74a473-4f23-45cc-883f-ed6ebe7167cf
1213
1217
# ╟─3691eca5-604c-4947-8b44-6f0b16798187
1214
1218
# ╟─6b93441f-e476-4c81-ad56-4df9222ade40
1215
-
#╠═58e61280-9add-40e1-b189-a4750591c804
1219
+
#╟─58e61280-9add-40e1-b189-a4750591c804
1216
1220
# ╟─01f2ceeb-9498-4c91-bd19-7f506752c005
1217
1221
# ╟─7fa5a486-0801-4977-8163-5a1fb02d59c9
1218
1222
# ╟─1de20b3e-5d93-4180-84d8-a2cbf2ab2966
1219
1223
# ╟─c9484783-f791-4c68-a2e0-89f2a71fe851
1220
-
#╠═373b57ee-68b2-4f2c-a94c-f06173c9ea2b
1224
+
#╟─373b57ee-68b2-4f2c-a94c-f06173c9ea2b
1221
1225
# ╟─4161b697-441e-4821-ba52-7f244886bd31
1222
1226
# ╟─df6c20bd-fcb9-458f-a592-39f5a3781aa1
1223
1227
# ╟─e0141ff8-e681-45cc-b9a8-6908ed332fef
@@ -1232,5 +1236,5 @@ Foldable(md"Hint...", md"In such _adversarial_ or _strategic_ environments, the
0 commit comments